add: demo app code and deps.

.gitignore

@@ -0,0 +1 @@
+__pycache__/

Dockerfile

@@ -0,0 +1,20 @@
+FROM redis/redis-stack:latest
+
+RUN apt-get update && apt-get install -y \
+  pip \
+  libxrender1 libxext6 \
+  && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir torch==2.6.0+cpu --index-url https://download.pytorch.org/whl/cpu
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY redis.conf .
+ENV REDIS_HOST=localhost
+ENV REDIS_PORT=6379
+
+COPY ./src ./src
+
+CMD ["sh", "-c", "redis-server redis.conf & python3 src/app.py"]

redis.conf

@@ -0,0 +1,2 @@
+loadmodule /opt/redis-stack/lib/redisearch.so
+maxmemory 768mb

requirements.txt

@@ -0,0 +1,11 @@
+chem-mrl>=0.5.8
+gradio[mcp]>=4.0.0
+transformers>=4.30.0
+torch>=2.0.0
+redis>=4.5.0
+redis-om>=0.2.0
+numpy>=1.24.0
+python-dotenv>=1.0.0
+scikit-learn>=1.3.0
+rdkit>=2025.3.2
+sentence-transformers==4.0.1

src/app.py

@@ -0,0 +1,218 @@
+import gradio as gr
+import numpy as np
+import pandas as pd
+from rdkit import Chem
+from rdkit.Chem import Draw
+from rdkit.Chem.Draw import rdMolDraw2D
+
+from constants import EMBEDDING_DIMENSION, LAUNCH_PARAMETERS, SUPPORTED_EMBEDDING_DIMENSIONS
+from data import SAMPLE_SMILES
+from service import MolecularEmbeddingService, SimilarMolecule, setup_logger
+
+logger = setup_logger()
+
+
+class App:
+    def __init__(self):
+        self.embedding_service = MolecularEmbeddingService()
+        self.demo = self.create_gradio_interface()
+
+    def molecule_similarity_search_pipeline(
+        self, smiles: str, embed_dim: int
+    ) -> tuple[list[float], list[SimilarMolecule], str]:
+        """Complete pipeline: SMILES -> Canonical SMILES -> Embedding -> Similar molecules"""
+        try:
+            if not smiles or smiles.strip() == "":
+                return [], [], "Please provide a valid SMILES string"
+
+            logger.info(f"Running similarity search: {smiles} - ({embed_dim})")
+            embedding = self.embedding_service.get_molecular_embedding(smiles, embed_dim)
+            neighbors = self.embedding_service.find_similar_molecules(embedding, embed_dim)
+
+            return embedding.tolist(), neighbors, "Search completed successfully"
+
+        except Exception as e:
+            error_msg = f"Search failed: {str(e)}"
+            logger.error(error_msg)
+            return [], [], error_msg
+
+    @staticmethod
+    def _truncated_attribute(obj, attr, max_len=45):
+        return f"{obj[attr][:max_len]}{'...' if len(obj[attr]) > max_len else ''}"
+
+    @classmethod
+    def _draw_molecule_grid(cls, similar: list[SimilarMolecule]) -> np.ndarray:
+        mols = [Chem.MolFromSmiles(m["smiles"]) for m in similar]
+        legends = [
+            f"{cls._truncated_attribute(m, 'name')}\n{m['category']}\n"
+            f"{cls._truncated_attribute(m, 'smiles')}\n({m['score']:.2E})"
+            for m in similar
+        ]
+
+        draw_options = rdMolDraw2D.MolDrawOptions()
+        draw_options.legendFontSize = 17
+        draw_options.legendFraction = 0.29
+        draw_options.drawMolsSameScale = False
+
+        img = Draw.MolsToGridImage(
+            mols,
+            legends=legends,
+            molsPerRow=2,
+            subImgSize=(250, 250),
+            drawOptions=draw_options,
+        )
+        return img
+
+    @staticmethod
+    def _display_sample_molecules(mols: pd.DataFrame):
+        for _, row in mols.iterrows():
+            with gr.Group():
+                gr.Textbox(value=row["smiles"], label=f"{row['name']} ({row['category']})", interactive=False, scale=3)
+                sample_btn = gr.Button(
+                    f"Load {row['name']}",
+                    scale=1,
+                    size="sm",
+                    variant="primary",
+                )
+                sample_btn.click(
+                    fn=None,
+                    js=f"() => {{window.setJSMESmiles('{row['smiles']}');}}",
+                )
+
+    @staticmethod
+    def clear_all():
+        return "", [], [], None, "Cleared - Draw a new molecule or enter SMILES"
+
+    def handle_search(self, smiles: str, embed_dim: int):
+        if not smiles.strip():
+            return (
+                [],
+                [],
+                None,
+                "Please draw a molecule or enter a SMILES string",
+            )
+        embedding, similar, status = self.molecule_similarity_search_pipeline(smiles, embed_dim)
+        img = self._draw_molecule_grid(similar)
+        return embedding, similar, img, status
+
+    def create_gradio_interface(self):
+        """Create the Gradio interface optimized for JavaScript client usage"""
+        head_scripts = """
+<link rel="preconnect" href="https://jsme-editor.github.io">
+<link rel="preload" href="https://jsme-editor.github.io/dist/jsme/jsme.nocache.js" as="script" crossorigin="anonymous">
+<link rel="preload" href="gradio_api/file=src/main.min.js" as="script">
+<script type="text/javascript" src="https://jsme-editor.github.io/dist/jsme/jsme.nocache.js" crossorigin="anonymous" defer></script>
+<script type="text/javascript" src="gradio_api/file=src/main.min.js" defer></script>
+        """
+
+        with gr.Blocks(
+            title="Chem-MRL: Molecular Similarity Search Demo",
+            theme=gr.themes.Soft(),
+            head=head_scripts,
+        ) as demo:
+            gr.Markdown("""
+            # 🧪 Chem-MRL: Molecular Similarity Search Demo
+
+            Use the JSME editor to draw a molecule or input a SMILES string.
+            The backend encodes the molecule using the Chem-MRL model to produce a vector embedding.
+            Similarity search is performed via an HNSW-indexed Redis vector store to retrieve closest matches.<br/>
+
+            [Model Repo](https://github.com/emapco/chem-mrl) | [Demo Repo](https://github.com/emapco/chem-mrl-demo)
+            """)
+            with gr.Tab("🔬 Molecular Search"), gr.Row():
+                with gr.Column(scale=1):
+                    gr.Markdown("### Molecule Input")
+                    gr.HTML("<div id='jsme_container'></div>")
+
+                    smiles_input = gr.Textbox(
+                        label="SMILES String",
+                        placeholder="Draw a molecule above or enter SMILES here (e.g., CCO for ethanol)",
+                        lines=2,
+                        elem_id="smiles_input",
+                    )
+
+                    embedding_dimension = gr.Dropdown(
+                        choices=SUPPORTED_EMBEDDING_DIMENSIONS,
+                        value=EMBEDDING_DIMENSION,
+                        label="Embedding Dimension",
+                        elem_id="embedding_dimension",
+                    )
+
+                    with gr.Row():
+                        search_btn = gr.Button(
+                            "🔍 Search Molecule Database",
+                            variant="primary",
+                            elem_id="search_btn",
+                        )
+                        clear_btn = gr.Button("🗑️ Clear All", variant="secondary")
+
+                with gr.Column(scale=1):
+                    gr.Markdown("### Search Results")
+                    status_output = gr.Textbox(
+                        label="Status",
+                        interactive=False,
+                        elem_id="status_output",
+                        value="Ready - Draw a molecule or enter SMILES",
+                    )
+
+                    with gr.Accordion("Molecular Embedding Vector", open=False):
+                        embedding_output = gr.JSON(
+                            label="Molecular Embedding",
+                            elem_id="embedding_output",
+                        )
+
+                    with gr.Accordion("Similar Molecules Response", open=False):
+                        similar_molecules_output = gr.JSON(
+                            label="API Response",
+                            elem_id="similar_molecules_output",
+                        )
+
+                    molecule_image = gr.Image(label="Similiar Molecules Grid", type="pil")
+
+            with gr.Tab("📊 Sample Molecules"):
+                gr.Markdown("""
+                ### Sample Molecules in Database
+                Click any button below to load the molecule into the JSME editor:
+                """)
+
+                with gr.Row():
+                    with gr.Column(scale=1):
+                        self._display_sample_molecules(SAMPLE_SMILES[::3])
+                    with gr.Column(scale=1):
+                        self._display_sample_molecules(SAMPLE_SMILES[1::3])
+                    with gr.Column(scale=1):
+                        self._display_sample_molecules(SAMPLE_SMILES[2::3])
+
+            search_btn.click(
+                fn=self.handle_search,
+                inputs=[smiles_input, embedding_dimension],
+                outputs=[
+                    embedding_output,
+                    similar_molecules_output,
+                    molecule_image,
+                    status_output,
+                ],
+                api_name="molecule_similarity_search_pipeline",
+            )
+
+            # Clear UI state
+            clear_btn.click(
+                fn=self.clear_all,
+                js="window.clearJSME",
+                outputs=[
+                    smiles_input,
+                    embedding_output,
+                    similar_molecules_output,
+                    molecule_image,
+                    status_output,
+                ],
+            )
+
+            gr.set_static_paths(paths=["src/"])
+
+        return demo
+
+
+if __name__ == "__main__":
+    app = App()
+    app.demo.launch(**LAUNCH_PARAMETERS)

src/constants.py

@@ -0,0 +1,33 @@
+# Model config
+MODEL_NAME = "Derify/ChemMRL-alpha"
+SUPPORTED_EMBEDDING_DIMENSIONS = [1024, 768, 512, 256, 128, 64, 32, 16, 8, 4, 2]
+EMBEDDING_DIMENSION = max(SUPPORTED_EMBEDDING_DIMENSIONS)
+
+# HNSW index parameters
+HNSW_K = 6
+HNSW_PARAMETERS = {
+    # Embedding vector dtype
+    "TYPE": "FLOAT32",
+    # Embedding vectors are normalized so COSINE and IP are equivalent
+    "DISTANCE_METRIC": "IP",
+    # Defines the initial capacity of the vector index. It helps in pre-allocating space for the index.
+    "INITIAL_CAP": 440,
+    # Max number of outgoing edges (connections) for each node in a graph layer.
+    "M": 32,
+    # Max number of connected neighbors to consider during graph building.
+    # Higher values increase accuracy, but also increase index build time.
+    "EF_CONSTRUCTION": 512,
+    # Max top candidates during KNN search. Higher values increase accuracy, but also increase search latency.
+    "EF_RUNTIME": 10,
+}
+
+# Gradio launch parameters
+LAUNCH_PARAMETERS = {
+    "server_name": "0.0.0.0",
+    "server_port": 7860,
+    "share": True,
+    "debug": False,
+    "show_api": False,
+    "pwa": True,
+    "mcp_server": False,
+}

src/data.py

@@ -0,0 +1,13 @@
+import os
+
+import pandas as pd
+
+__data_dir = os.path.join(os.path.dirname(__file__), "data")
+
+__dataset_smiles_file = os.path.join(__data_dir, "dataset_smiles.csv")
+__sample_smiles_file = os.path.join(__data_dir, "sample_smiles.csv")
+__isomer_design_subset_file = os.path.join(__data_dir, "isomer_design_subset.csv")
+
+DATASET_SMILES = pd.read_csv(__dataset_smiles_file)
+SAMPLE_SMILES = pd.read_csv(__sample_smiles_file)
+ISOMER_DESIGN_SUBSET = pd.read_csv(__isomer_design_subset_file)

src/data/dataset_smiles.csv

@@ -0,0 +1,22 @@
+smiles,name,category
+CCO,Ethanol,Alcohol
+CC(=O)O,Acetic acid,Carboxylic acid
+c1ccccc1,Benzene,Aromatic hydrocarbon
+CC(C)O,Isopropanol,Alcohol
+CCN(CC)CC,Triethylamine,Amine
+c1ccc(cc1)O,Phenol,Aromatic alcohol
+CC(=O)OC1=CC=CC=C1C(=O)O,Aspirin,Carboxylic acid ester
+CN1C=NC2=C1C(=O)N(C(=O)N2C)C,Caffeine,Alkaloid
+CC(C)(C)OC(=O)NC1CCC(CC1)O,Boc-protected cyclohexanol,Protected alcohol
+CCCCCCCCCCCCCCC(=O)O,Palmitic acid,Fatty acid
+c1ccc2c(c1)cccn2,Quinoline,Aromatic heterocycle
+CC1=CC=C(C=C1)C,p-Xylene,Aromatic hydrocarbon
+CCCCO,Butanol,Alcohol
+CC(C)C,Isobutane,Alkane
+c1ccc(cc1)N,Aniline,Aromatic amine
+CC(=O)N,Acetamide,Amide
+CCCCCCCCCCCCCCCCCC(=O)O,Stearic acid,Fatty acid
+c1ccc(cc1)C(=O)O,Benzoic acid,Aromatic carboxylic acid
+CCc1ccccc1,Ethylbenzene,Aromatic hydrocarbon
+CC(C)CC(C)(C)C,"2,2,4-Trimethylpentane",Branched alkane
+O=C1c2ccccc2C(=O)N1C3CCC(=O)NC3=O,Thalidomide,Phthalimides

src/data/isomer_design_subset.csv

@@ -0,0 +1,415 @@
+smiles,name,category
+O=C1CCCCC1(c1ccccc1Cl)N1CCCCC1,2-Cl-2'-Oxo-PCP,arylcycloalkylamine
+CCCCNC1(c2ccccc2)CCCCC1,PCB,arylcycloalkylamine
+COc1ccccc1C1(N2CCCCC2)CCCCC1,2-MeO-PCP,arylcycloalkylamine
+c1ccc(C2(n3cccc3)CCCCC2)cc1,PCPyrrole,arylcycloalkylamine
+Fc1cccc(C2(N3CCCC3)CC3CCC2C3)c1,3F-PBCHPy,arylcycloalkylamine
+COc1ccc(C2(NCC3CCCO3)CCCCC2)cc1,1-(4-Methoxyphenyl)-N-[(oxolan-2-yl)methyl]cyclohexan-1-amine,arylcycloalkylamine
+COc1ccc(C2(N3CCCC3)CC3CCC2C3)cc1,4-MeO-PBCHPy,arylcycloalkylamine
+C=CCN(CC)C1(c2ccccc2)CCCCC1,N-Allyl-PCE,arylcycloalkylamine
+CCCCCNC1(c2ccccc2)CCCCC1,PCPe,arylcycloalkylamine
+OC1CCCCC1(c1ccccc1)N1CCCCC1,2'-HO-PCP,arylcycloalkylamine
+Cc1ccccc1C1(N2CCCC(O)C2)CCCCC1,3″-HO-2-Me-PCP,arylcycloalkylamine
+c1ccc(C2(N3CCCCC3)CCCC2)cc1,PCPEP,arylcycloalkylamine
+CCN(C)C1(c2ccccc2C)CCCCC1,2-Me-PCEM,arylcycloalkylamine
+Fc1ccccc1C1(N2CCCC2)CC2CCC1C2,2F-PBCHPy,arylcycloalkylamine
+CC(C)OCCCNC1(c2ccccc2)CCCCC1,1-Phenyl-N-{3-[(propan-2-yl)oxy]propyl}cyclohexan-1-amine,arylcycloalkylamine
+CCNC1(c2ccc(C)cc2)CCCCC1,4-Me-PCE,arylcycloalkylamine
+CCNC1(c2cccc(OC)c2)CCCCC1=O,MXE,arylcycloalkylamine
+CCOC(=O)CCCN(C)C1(c2ccccc2Cl)CCCCC1=O,Et-KE-Bu,arylcycloalkylamine
+CCNC1(c2ccccc2)CCCCC1,PCE,arylcycloalkylamine
+c1cc2c(cc1C1(N3CCOCC3)CCCCC1)OCO2,"3,4-MD-PCMo",arylcycloalkylamine
+OCCNC1(c2ccccc2)CCCCC1,PCEOH,arylcycloalkylamine
+c1ccc(C2(N3CCCCC3)CCCCCCC2)cc1,PCOP,arylcycloalkylamine
+COc1ccc(C2(N3CCCCC3)CC3CCC2C3)cc1,4-MeO-PBCHP,arylcycloalkylamine
+COc1cccc(C2(N3CCCCC3)CCCc3ccccc32)c1,"1-[1-(3-Methoxyphenyl)-1,2,3,4-tetrahydronaphthalen-1-yl]piperidine",arylcycloalkylamine
+CCNC12CCCCC1CCc1ccccc12,"N-Ethyl-1,3,4,9,10,10a-hexahydrophenanthren-4a(2H)-amine",arylcycloalkylamine
+c1ccc([C@]2(N3CCCCC3)C[C@H]3C[C@H]3C2)cc1,"1-[(1R,3r,5S)-3-Phenylbicyclo[3.1.0]hexan-3-yl]piperidine",arylcycloalkylamine
+c1ccc(C2(N3CCCCC3)CN3CCC2CC3)cc1,P3QP,arylcycloalkylamine
+COc1cccc(C2(N3CCCC3)CCCCC2=O)c1,3-MeO-2'-Oxo-PCPy,arylcycloalkylamine
+c1cc2c(cc1C1(N3CCCCC3)CCCCC1)OCCO2,"3,4-ED-PCP",arylcycloalkylamine
+NC1(c2ccccc2Cl)CCCNC1=O,3-Amino-3-(2-chlorophenyl)piperidin-2-one,arylcycloalkylamine
+CC[C@]1(c2ccccc2)OC[C@H]([C@@H]2CCCCN2)O1,CL-1848C,arylcycloalkylamine
+O=[N+]([O-])c1cccc(C2(N3CCCCC3)CCCCC2)c1,3-NO2-PCP,arylcycloalkylamine
+Fc1cccc(C2(N3CCCCC3)CCCCCC2)c1,3F-PCHEP,arylcycloalkylamine
+CCNC1(c2cccc(OC(C)C)c2)CCCCC1=O,iPXE,arylcycloalkylamine
+COc1cc(C2(N3CCCCC3)CCCCC2)ccc1C,3-MeO-4-Me-PCP,arylcycloalkylamine
+O=C1CCC(c2ccccc2)(N2CCCCC2)CC1,4'-Oxo-PCP,arylcycloalkylamine
+CCNC1(c2ccc3c(c2)OCO3)CCCCC1,"3,4-MD-PCE",arylcycloalkylamine
+CNC1(c2ccccc2OC)CCCCC1=O,2-MDCK,arylcycloalkylamine
+CCOc1cccc(C2(N3CCCCC3)CCCCC2)c1,3-EtO-PCP,arylcycloalkylamine
+CNC1(c2ccccc2C)CCCCC1=O,o-MDCK,arylcycloalkylamine
+CNC1(c2cccc(Cl)c2)CCCCC1=O,meta-Ketamine,arylcycloalkylamine
+CCNC1(c2cc(OC)ccc2Br)CCCCC1=O,Bromo-methoxetamine,arylcycloalkylamine
+CNC1(c2cccc(OC)c2)CCCCC1=O,MXM,arylcycloalkylamine
+CCCOC(=O)CCNC1(c2ccccc2Cl)CCCCC1=O,nPr-NKE-Et,arylcycloalkylamine
+CC1CCCCC1(c1ccccc1)N1CCCCC1,2'-Me-PCP,arylcycloalkylamine
+Cc1ccc(C2(N3CCC(O)CC3)CCCCC2O)cc1,"2',4″-HO-4-Me-PCP",arylcycloalkylamine
+NC1(c2ccccc2Cl)CCC(O)CC1=O,"(2,5)-HNK",arylcycloalkylamine
+Fc1cccc(C2(N3CCCCC3)CC3CCC2C3)c1,3F-PBCHP,arylcycloalkylamine
+CC(C)NC1(c2ccccc2)CCCCC1,PCiP,arylcycloalkylamine
+Fc1ccc(C2(N3CCCCC3)CCCCCC2)cc1,4F-PCHEP,arylcycloalkylamine
+CCCNC1(c2ccccc2)CCCC1,PCPEPr,arylcycloalkylamine
+CC(C)Oc1cccc(C2(N)CCCCC2)c1,3-IpO-PCA,arylcycloalkylamine
+CC(O)CNC1(c2ccccc2)CCCCC1,1-[(1-Phenylcyclohexyl)amino]propan-2-ol,arylcycloalkylamine
+Cc1ccc(C2(N(C)C)CCC(O)(CCc3ccccc3)CC2)cc1,MDPC,arylcycloalkylamine
+Cc1cc(O)ccc1C1(N2CCOCC2)CCCCC1,4-HO-2-Me-PCMo,arylcycloalkylamine
+O=C1CCCCC1(c1ccccc1)N1CCCC1,2-Phenyl-2-(pyrrolidin-1-yl)cyclohexan-1-one,arylcycloalkylamine
+CNC1(c2ccccc2Cl)CCCCCC1=O,2-(2-Chlorophenyl)-2-(methylamino)cycloheptan-1-one,arylcycloalkylamine
+CCN(CC)C1(c2ccccc2)CCCCC1,PCDE,arylcycloalkylamine
+COc1cccc([C@@]2(O)CCN(C3(c4ccccc4)CCCCC3)C[C@@H]2CN(C)C)c1,US 7049327 #6,arylcycloalkylamine
+Cc1cccc(C2(N3CCCC3)CC3CCC2C3)c1,3-Me-PBCHPy,arylcycloalkylamine
+C#Cc1ccc2c(c1)C(c1ccccc1F)=N[C@@H](C)c1c(C(=O)N(C)C)ncn1-2,GL-i-54,aryldiazepine
+O=C1CN=C(c2c(F)cccc2F)c2cc(Cl)ccc2N1,Ro-07-3953,aryldiazepine
+O=C1CN=C(c2ccn[nH]2)c2cc(Cl)ccc2N1,SCHEMBL9684958,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccn1)=NCc1c(C(N)=O)ncn1-2,MP-ii-065,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2c(Cl)cccc21,Ro-13-0699,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccc1)=NCc1c(C(=O)OCC(F)(F)F)ncn1-2,DM-ii-20,aryldiazepine
+O=C1OCCC2N=C(c3ccccc3)c3ccccc3-n3cnc1c32,"8-Phenyl-6,6a-dihydro-4-oxa-2,7,12b-triazabenzo[b]cyclopenta[kl]heptalen-3(5H)-one",aryldiazepine
+CCC1C(C)=NN=C(c2ccc(OC)c(OC)c2)c2cc(OC)c(OC)cc21,Tofisopam,aryldiazepine
+CN1C(=O)CN=C(c2ccc(F)cc2)c2cc(Cl)ccc21,4F-Diazepam,aryldiazepine
+CC(C)COC(=O)c1ncn2c1C1CCCN1C(=O)c1c(Cl)cccc1-2,Ro-16-7082,aryldiazepine
+CN1C(=O)CN=C(c2cccs2)c2cc(Cl)ccc21,WZ-113,aryldiazepine
+CC(=O)c1ccc2c(c1)C(c1ccccc1)=NCC(=O)N2,Ro-20-3053,aryldiazepine
+C#Cc1ccc2c(c1)C(=O)N(C)Cc1c(C(=O)OC(C)(C)C)ncn1-2,RY-24,aryldiazepine
+CC(C)(C)OC(=O)c1ncn2c1C1CCCN1C(=O)c1c(F)cccc1-2,Ro-16-8912,aryldiazepine
+O=C1CN=C(c2ccccc2F)c2cc(Cl)ccc2N1CCO,Ro-07-2750,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2Cl)c2cc([N+](=O)[O-])ccc21,Ro-05-4082,aryldiazepine
+Cc1nnc2n1-c1sc(Cl)cc1C(c1ccccc1)=NC2,Deschloroclotizolam,aryldiazepine
+CCOC(=O)c1ncn2c1C(C)N=C(c1ccc(F)cc1)c1cc(F)ccc1-2,"Ethyl 8-fluoro-6-(4-fluorophenyl)-4-methyl-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate",aryldiazepine
+O=C1CN=C(c2ccccc2F)c2cc(Cl)c(Cl)cc2N1,Ro-20-8065,aryldiazepine
+CN1C(=O)CN=C(c2c(F)cccc2F)c2cc(Br)ccc21,Ro-13-3780,aryldiazepine
+CN1C(=O)CC(=O)N(c2ccccc2)c2cc(Cl)ccc21,Clobazam,aryldiazepine
+OCc1nnc2n1-c1ccc(Cl)cc1C(c1ccccc1)=NC2,alpha-Hydroxyalprazolam,aryldiazepine
+O=C1Nc2ccc(Cl)cc2C(c2ccccc2)=NC1O,Oxazepam,aryldiazepine
+CCOC(=O)c1ncn2c1CN=C(c1ccc([N+](=O)[O-])cc1)c1ccccc1-2,Ro-23-2896,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2cc(Br)ccc21,Flubrometazepam,aryldiazepine
+CN(C)C(=O)c1nc2n(n1)-c1ccc(Cl)cc1C(c1ccccc1Cl)=NC2,Rilmazolam,aryldiazepine
+CC1N=C(c2ccccc2F)c2cc(Cl)ccc2NC1=O,Ro-11-4878,aryldiazepine
+CCOC(=O)c1ncn2c1C1CCCN1C(=O)c1cc(Br)ccc1-2,RY-054,aryldiazepine
+CCOC(=O)c1ncn2c1C1CCN1C(=O)c1c(Cl)cccc1-2,Ro-16-0858,aryldiazepine
+CN1C(=O)CN=C(c2ccc(Cl)cc2)c2cc(Cl)ccc21,Ro-05-4864,aryldiazepine
+CN(C)CCc1nnc2n1-c1ccc(Cl)cc1C(c1ccccc1)=NC2,U-43465F,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2c(Cl)cc(Cl)cc21,Ro-13-0882,aryldiazepine
+CCc1ccc2c(c1)C(c1ccccc1)=NCC(=O)N2,Ro-20-2533,aryldiazepine
+CCOC(=O)c1ncn2c1CC=C(c1ccccc1F)c1ccccc1-2,Ro-15-2200,aryldiazepine
+O=C1CC=C(c2ccccc2Cl)c2cc([N+](=O)[O-])ccc2N1,Ro-15-8867,aryldiazepine
+Cc1cc2c(cc1Cl)NC(=O)CN=C2c1ccccc1F,Ro-20-8552,aryldiazepine
+N#Cc1ncn2c1CN=C(c1ccc(Cl)cc1)c1cc(Cl)ccc1-2,"8-Chloro-6-(4-chlorophenyl)-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carbonitrile",aryldiazepine
+CC(C)c1nc(-c2ncn3c2CN(C)C(=O)c2c(Cl)cccc2-3)no1,FG-8205,aryldiazepine
+Cc1nnc2n1-c1ccc([N+](=O)[O-])cc1C(c1ccccc1Cl)=NC2,Clonazolam,aryldiazepine
+O=C1CN=C(c2ccccc2[N+](=O)[O-])c2cc(Br)ccc2N1,QH-II-067a,aryldiazepine
+NC(=O)c1ncn2c1CC=C(c1ccccc1F)c1cc(Cl)ccc1-2,Ro-14-3930,aryldiazepine
+COC(=O)CC[C@@H]1N=C(c2ccccn2)c2cc(Br)ccc2-n2c(C)cnc21,CNS-7056,aryldiazepine
+CN1C(=O)CN=C(C2CCC2)c2cc(F)ccc21,"5-Cyclobutyl-7-fluoro-1-methyl-1,3-dihydro-2H-1,4-benzodiazepin-2-one",aryldiazepine
+O=C1CN=C(c2c(F)cccc2Cl)c2cc(Cl)ccc2N1,Ro-07-5193,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2cc(I)ccc21,Ro-07-9957,aryldiazepine
+O=C1CN=C(c2ccccc2)c2cc(Cl)ccc2N1CC1CC1,Prazepam,aryldiazepine
+CCOC(=O)c1ncn2c1CN=C(c1cccc([N+](=O)[O-])c1)c1cc(F)ccc1-2,"Ethyl 8-fluoro-6-(3-nitrophenyl)-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate",aryldiazepine
+Clc1ccc2c(c1)-c1ncnn1Cc1c(Cl)ncn1-2,Ro4882224,aryldiazepine
+[N-]=[N+]=Nc1ccc2c(c1)C(c1ccccc1F)=NCC(=O)N2,Ro-14-3074,aryldiazepine
+Fc1ccccc1C1=NCC=Cc2ccc(Cl)cc21,Ro-22-1274,aryldiazepine
+Nc1ncc2c(n1)-c1ccc(Cl)cc1C(c1ccccc1F)=NC2,Ro-22-2038,aryldiazepine
+Clc1ccc2c(c1)C(c1ccccc1)=NCc1nncn1-2,Estazolam,aryldiazepine
+Cc1nn(C)c2c1C(c1ccccc1F)=NCC(=O)N2C,Zolazepam,aryldiazepine
+Cc1nnc2n1-c1ccc(Cl)cc1C(c1ccccn1)=NC2,Pyclazolam,aryldiazepine
+S=C1CN=C(c2ccccc2)c2cc(Cl)ccc2N1,Thionordazepam,aryldiazepine
+O=C1CN=C(c2ccccc2)c2cc(Cl)ccc2N1CC(F)(F)F,Halazepam,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccc1)=NCC1=N/C(=C\N3CCN(C)CC3)C(=O)N12,PS-i-37,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccc1F)=NCc1c(C(=O)OCC(F)(F)F)ncn1-2,JY-i-59,aryldiazepine
+O=C1CN=C(c2ccccc2F)c2cc([N+](=O)[O-])ccc2N1,Fonazepam,aryldiazepine
+CN1CCN(/C=C2\N=C3CN=C(c4ccccc4)c4cc(Br)ccc4N3C2=O)CC1,PS-1-35,aryldiazepine
+CCCCCCC(C)(C)c1ccc([C@@H]2CCC[C@H](O)C2)c(O)c1,"(1R,3S) CP 47,497",cannabinoid
+CCCc1ccc(C(=O)c2cn(CCCCCF)c3ccccc23)c2ccccc12,SGT-62,cannabinoid
+CC(C)[C@H](NC(=O)c1cn(Cc2ccc(F)cc2)c2ccccc12)C(N)=O,AB-FUBICA,cannabinoid
+CCCCCn1nc(C(=O)N[C@@H](Cc2ccccc2)C(=O)OC)c2ccccc21,MPP-PINACA,cannabinoid
+O=C(Cc1cn(Cc2ccc(F)cc2)c2ccccc12)NC1C2CC3CC(C2)CC1C3,AFUBIATA,cannabinoid
+CC(C)(C)c1nc(N2CCC(F)(F)C2)c2nnn(Cc3ccccc3S(C)(=O)=O)c2n1,"5-tert-Butyl-7-(3,3-difluoropyrrolidin-1-yl)-3-{[2-(methanesulfonyl)phenyl]methyl}-3H-[1,2,3]triazolo[4,5-d]pyrimidine",cannabinoid
+CCC(C(N)=O)N(C)C(=O)c1nn(Cc2ccc(F)cc2)c2ccccc12,AB-FUBINACA isomer 5,cannabinoid
+COc1cccc2c3c(=O)n([C@@H]4C(C)(C)[C@@H]5CC[C@@]4(C)C5)ccc3n(Cc3ccncc3)c12,WO 2001/058869 #190,cannabinoid
+CCCC(F)Cn1cc(C(=O)C2C(C)(C)C2(C)C)c2ccccc21,(2-Fluoropentyl) XLR-11,cannabinoid
+O=C(Nc1ccccc1)c1cn(CCCCCF)c2ccccc12,LTI-701,cannabinoid
+C=CCCCn1nc(C(=O)N[C@H](C(=O)OC)C(C)C)c2ccccc21,MMB-4en-PINACA,cannabinoid
+O=C(Nc1ccccc1)c1nn(CC2CCOCC2)c2ccccc12,SGT-224,cannabinoid
+Cc1ccc(C(=O)NC(C)(C)c2ccccc2)cc1S(=O)(=O)N1CCC(F)(F)CC1,SGT-233,cannabinoid
+CCCCCn1nc(C(=O)c2ccc(OC)c3ccccc23)c2ccccc21,SGT-35,cannabinoid
+CC1(C)C(C(=O)c2cn(CC3CCCCC3)c3ccccc23)C1(C)C,(CHM) UR-144,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2ccc(N=C=S)c3ccccc23)c2ccc(N=C=S)cc21,AM-2215,cannabinoid
+CCCCCCC1(c2ccc(-c3cc(C)cc(C)c3)c(O)c2)CC1,"4-(1-Hexylcyclopropyl)-3',5'-dimethyl[1,1'-biphenyl]-2-ol",cannabinoid
+COC(=O)[C@@H](NC(=O)c1cn(Cc2ccc(F)cc2)c2ncccc12)C(C)(C)C,MDMB-FUB7AICA,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2ccccc2I)c2ccccc21,AM-2233,cannabinoid
+O=C(c1cccc2ccccc12)n1c(-c2ccncc2)nc2ccccc21,(Naphthalen-1-yl)[2-(pyridin-4-yl)-1H-benzimidazol-1-yl]methanone,cannabinoid
+C=CCCCn1nc(C(=O)N[C@H](C(=O)OC)C(C)(C)C)c2ccccc21,MDMB-4en-PINACA,cannabinoid
+Oc1ccc(/C=C2\C=C(CCN3CCOCC3)c3ccccc32)c2ccccc12,US 5292736 #19,cannabinoid
+CCc1c(C(=O)NN2CCCCC2)nn(-c2ccc(Cl)cc2Cl)c1-c1ccc(C#Cc2ccc(C(F)(F)F)cc2)s1,TM-38837,cannabinoid
+COC(=O)[C@H](Cc1ccccc1)NC(=O)c1cn(CCCCCF)c2ccccc12,MPHP-2201,cannabinoid
+Cc1ccc2c(C(C[N+](=O)[O-])c3ccccc3)c(-c3ccccc3)[nH]c2c1,GAT211,cannabinoid
+CCCCCn1nc(C(=O)NC2CCCCC2)c2ccccc21,SGT-85,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2ccc([N+](=O)[O-])c3ccccc23)c2ccccc21,AM-1299,cannabinoid
+CCc1c(C)cc(C(=O)NC2(C(=O)O)CCCCC2)c(=O)n1Cc1ccc(F)cc1,S-777469,cannabinoid
+CCCCCC(C)NC(=O)c1nn(CCCCC)c2ccccc12,SGT-71,cannabinoid
+CC1(C)Oc2cc(C34CC5CC(CC(C5)C3)C4)cc(O)c2[C@@H]2CC(O)CC[C@H]21,"(6aR,10aR)-3-(Adamantan-1-yl)-6,6-dimethyl-6a,7,8,9,10,10a-hexahydro-6H-dibenzo[b,d]pyran-1,9-diol",cannabinoid
+C=C(C)[C@H]1CCC(CO)=C[C@@H]1c1c(O)cc(C(C)(C)CCCCN2CCOCC2)cc1O,"(1'S,2'S)-5'-(Hydroxymethyl)-4-[2-methyl-6-(morpholin-4-yl)hexan-2-yl]-2'-(prop-1-en-2-yl)-1',2',3',4'-tetrahydro[1,1'-biphenyl]-2,6-diol",cannabinoid
+O=C(N/N=C1\C(=O)N(CCCCCF)c2ccccc21)c1ccccc1,BZO-5F-POXIZID,cannabinoid
+CCCCCOc1c(OC)ccc2cc(C(=O)NCc3ccc4c(c3)OCO4)c(=O)[nH]c12,JTE-907,cannabinoid
+CCCCn1nc(C(=O)N[C@@H](Cc2ccccc2)C(=O)OC)c2ccccc21,MPP(N)-073,cannabinoid
+CCCCCCC(C)(C)c1ccc([C@@H]2CCC[C@@H](O)C2)c(O)c1,"(R,R)-epi-CP 47,497",cannabinoid
+CCCCCOc1ccc(C(=O)c2cccc3ccccc23)c2ccccc12,CB-13,cannabinoid
+COc1ccccc1OC(=O)c1ccc(C)c(S(=O)(=O)N2CCOCC2)c1,2-Methoxyphenyl 4-methyl-3-(morpholine-4-sulfonyl)benzoate,cannabinoid
+CCCCCC#Cc1cc(O)c2c(c1)OC(C)(C)[C@@H]1CC=C(C)C[C@@H]21,AMG-1,cannabinoid
+CC(C)[C@H](NC(=O)c1cn(CCCCCF)c2ccccc12)C(N)=O,5F-ABICA,cannabinoid
+CCCCCn1cc(C(=O)NC23CC4CC(CC(C4)C2)C3)c(=O)c2ccc(F)cc21,"N-(Adamantan-1-yl)-7-fluoro-4-oxo-1-pentyl-1,4-dihydroquinoline-3-carboxamide",cannabinoid
+CC1=CC[C@@H]2[C@@H](C1)c1c(O)cc(C(C)(C)CCCCCC#N)cc1OC2(C)C,O-774,cannabinoid
+CC(C)(C)[C@H](NC(=O)n1c(=O)n(CCN2CCOCC2)c2ccccc21)C(N)=O,WO 2008/032164 #4,cannabinoid
+COC(=O)[C@H](Cc1ccccc1)NC(=O)c1cn(CC2CCCCC2)c2ccccc12,MPP-CHMICA,cannabinoid
+Fc1ccc2nc(-c3ccccn3)n(Cc3cccc4ccccc34)c2c1,6-Fluoro-1-[(naphthalen-1-yl)methyl]-2-(pyridin-2-yl)-1H-benzimidazole,cannabinoid
+CCCCCn1cc(C(=O)N[C@@H](C)c2ccccc2)c2ccccc21,1-Pentyl-N-[(1S)-1-phenylethyl]-1H-indole-3-carboxamide,cannabinoid
+CCCCCCCC(C)c1cc(OC(=O)CCCN(CC)CC)c2c(c1)OC(C)(C)C1=C2CC(C)CC1,SP-325,cannabinoid
+CCCCCn1cc(C(=O)c2cccc3ccccc23)cc1-c1cccc(C)c1,JWH-346,cannabinoid
+O=C(Cc1ccccc1)N/N=C1\C(=O)N(CC2CCCCC2)c2cc(OCCN3CCOCC3)ccc21,"N'-{1-(Cyclohexylmethyl)-6-[2-(morpholin-4-yl)ethoxy]-2-oxo-1,2-dihydro-3H-indol-3-ylidene}-2-phenylacetohydrazide",cannabinoid
+Cc1c(C(=O)C2C(C)(C)C2(C)C)c2ccccc2n1CCCCCF,2-Me-XLR-11,cannabinoid
+COc1ccc2c(c1)CCC2CNC(=O)c1cn(CCCCCF)c2ccccc12,SGT-105,cannabinoid
+CCCCCc1cc(O)c2c(c1)OC(C)(C)C(CCC(C)=O)C2=O,CBCN,cannabinoid
+CCCCCn1nc(C(=O)NC(C)(C)c2ccc(F)cc2)c2ccccc21,SGT-60,cannabinoid
+CC(C)(C)[C@H](NC(=O)c1nn(CC2CCCCCC2)c2ccccc12)C(=O)NCCO,WO 2009/106980 #433,cannabinoid
+CC(C)(NC(=O)c1cn(Cc2ccc(F)cc2)c2ccccc12)c1ccccc1,CUMYL-FUBICA,cannabinoid
+COC(=O)[C@@H](NC(=O)c1nn(CC2CCCCC2)c2ccccc12)C(C)(C)C,MDMB(N)-CHM,cannabinoid
+CCCCCn1nc(C(=O)c2cccc3ccccc23)c2ccccc21,SGT-17,cannabinoid
+CCCCCn1cc(C(=O)N[C@H](C(N)=O)C(C)(C)C)c2ccccc21,ADBICA,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2cc(N=C=S)ccc2I)c2ccc(N=C=S)cc21,AM-2223,cannabinoid
+CCCCCCCCCn1cc(C(=O)c2cccc3ccccc23)c2ccccc21,(Nonyl) JWH-018,cannabinoid
+CCCCCCC(C)(C)c1cc(OC)c([C@@H]2C=C(CO)C3CC2C3(C)C)c(OC)c1,HU-433,cannabinoid
+C=CCc1cc(OC)c(OC)c(OC)c1OC,"2,3,4,5-Tetramethoxyphenylprop-2-ene",essential oil
+C/C=C\c1ccc(O)c(OC)c1,cis-4-Hydroxy-3-methoxyphenylprop-1-ene,essential oil
+C/C=C/c1ccc2c(c1OC)OCO2,Isocroweacin,essential oil
+C/C=C\c1cc(OC)c(O)c(OC)c1,"cis-4-Hydroxy-3,5-dimethoxyphenylprop-1-ene",essential oil
+C=CCc1ccc2c(c1OC)OCO2,Croweacin,essential oil
+C/C=C\c1cc2c(cc1OC)OCO2,"cis-2-Methoxy-4,5-methylenedioxyphenylprop-1-ene",essential oil
+C=CCc1ccc2c(c1)OCO2,Safrole,essential oil
+C/C=C\c1cc(OC)c2c(c1)OCO2,cis-Isomyristicin,essential oil
+C/C=C\c1ccc(OC)cc1OC,"cis-2,4-Dimethoxyphenylprop-1-ene",essential oil
+C/C=C\c1ccc2c(c1OC)OCO2,"cis-2-Methoxy-3,4-methylenedioxyphenylprop-1-ene",essential oil
+C/C=C/c1ccc(OC)c(OC)c1,Methylisoeugenol,essential oil
+C=CCc1ccc(OC)cc1OC,Osmorrhizole,essential oil
+C=CCc1ccc(OC)c(OC)c1,Methyleugenol,essential oil
+C=CCc1ccc(O)c(OC)c1,Eugenol,essential oil
+C=CCc1cc2c(c(OC)c1OC)OCO2,Dillapiole,essential oil
+C/C=C\c1ccc2c(c1)OCO2,alpha-Isosafrole,essential oil
+C/C=C/c1cc(OC)c(OC)c(OC)c1,Isoelemicin,essential oil
+C=CCc1ccc(OC)c(O)c1,Chavibetol,essential oil
+C=CCc1cc(OC)c(OC)c(OC)c1,Elemicin,essential oil
+C=CCc1cc(OC)cc(OC)c1,"3,5-Dimethoxyphenylprop-2-ene",essential oil
+C/C=C\c1cc(OC)c(OC)c(OC)c1,cis-Isoelemicin,essential oil
+C/C=C/c1cc(OC)c(OC)c(OC)c1OC,"trans-2,3,4,5-Tetramethoxyphenylprop-1-ene",essential oil
+C/C=C/c1cc(OC)c2c(c1OC)OCO2,Isoapiole,essential oil
+C/C=C\c1ccc(OC)c(O)c1,cis-3-Hydroxy-4-methoxyphenylprop-1-ene,essential oil
+C=CCc1ccc(OC)cc1,Estragole,essential oil
+C/C=C\c1cc(OC)c2c(c1OC)OCO2,"cis-2,5-Dimethoxy-3,4-methylenedioxyphenylprop-1-ene",essential oil
+C=CCc1cc(OC)c2c(c1OC)OCO2,Apiole,essential oil
+C/C=C/c1ccc(OC)cc1OC,Nothosmyrnol,essential oil
+C/C=C/c1cc(OC)cc(OC)c1,"trans-3,5-Dimethoxyphenylprop-1-ene",essential oil
+C/C=C\c1cc2c(c(OC)c1OC)OCO2,"cis-2,3-Dimethoxy-4,5-methylenedioxyphenylprop-1-ene",essential oil
+C/C=C\c1ccc(OC)cc1,cis-4-Methoxyphenylprop-1-ene,essential oil
+C/C=C/c1cc(OC)c(O)c(OC)c1,"2,6-Dimethoxy-4-[(1E)-prop-1-en-1-yl]phenol",essential oil
+C=CCc1cc(OC)c(OC)cc1OC,γ-Asarone,essential oil
+C/C=C/c1ccc(OC)cc1,Anethole,essential oil
+C=CCc1cc2c(cc1OC)OCO2,Asaricin,essential oil
+C/C=C/c1cc2c(c(OC)c1OC)OCO2,Isodillapiole,essential oil
+C/C=C/c1cc(OC)c(OC)cc1OC,Asarone,essential oil
+C/C=C/c1ccc(O)c(OC)c1,Isoeugenol,essential oil
+C/C=C\c1cc(OC)cc(OC)c1,"cis-3,5-Dimethoxyphenylprop-1-ene",essential oil
+C/C=C\c1cc(OC)c(OC)cc1OC,beta-Asarone,essential oil
+C=CCc1cc(OC)c2c(c1)OCO2,Myristicin,essential oil
+C/C=C/c1ccc(OC)c(O)c1,2-Methoxy-5-[(1E)-prop-1-en-1-yl]phenol,essential oil
+C/C=C/c1cc(OC)c2c(c1)OCO2,Isomyristicin,essential oil
+C/C=C\c1ccc(OC)c(OC)c1,"cis-3,4-Dimethoxyphenylprop-1-ene",essential oil
+C=CCc1cc(OC)c(O)c(OC)c1,5-Methoxyeugenol,essential oil
+C/C=C/c1ccc2c(c1)OCO2,beta-Isosafrole,essential oil
+C/C=C\c1cc(OC)c(OC)c(OC)c1OC,"cis-2,3,4,5-Tetramethoxyphenylprop-1-ene",essential oil
+C/C=C/c1cc2c(cc1OC)OCO2,Carpacin,essential oil
+CCC(=O)N(c1ccccc1)C1(COC)CCN(CC(O)c2cccs2)CC1,beta-Hydroxy-sufentanil,fentanyl
+CCC(=O)N(CCc1ccccc1)C1CCN(CCc2ccccc2)CC1,(ethylene) Fentanyl,fentanyl
+CCC(=O)N(c1ccccc1F)C1(c2ccccc2)CCN(CCn2cccn2)CC1,N-(2-Fluorophenyl)-N-{4-phenyl-1-[2-(1H-pyrazol-1-yl)ethyl]piperidin-4-yl}propanamide,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(C(C)Cc2cc(OC)c(I)cc2OC)CC1,N-(DOI) fentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1(COC(=O)C(C)(C)C)CCN(CCN2C(=O)c3ccccc3C2=O)CC1,"{1-[2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)ethyl]-4-[phenyl(propanoyl)amino]piperidin-4-yl}methyl 2,2-dimethylpropanoate",fentanyl
+CC(C)C(=O)N(c1cccc(F)c1)C1CCN(CCc2ccccc2)CC1,m-Fluoro-isobutyrylfentanyl,fentanyl
+CCCC(=O)N(c1ccc(OC)cc1)C1CCN(CCc2ccccc2)CC1,p-MeO-BF,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(Cc2ccccc2)CC1,Benzylcarfentanil,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCc2ccsc2)CC1,Methyl 4-[phenyl(propanoyl)amino]-1-[2-(thiophen-3-yl)ethyl]piperidine-4-carboxylate,fentanyl
+COCC(=O)N(c1ccc(C)cc1)C1CCN(CCc2ccccc2)CC1,p-Methyl-methoxyacetylfentanyl,fentanyl
+CCCCC(=O)N(c1ccc(OC)cc1)C1CCN(CCc2ccccc2)CC1,p-Methoxy-valerylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)CC1F,NFEPP,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(C(C)Cc2cc(OC)c(Br)cc2OC)CC1,N-(DOB) fentanyl,fentanyl
+CCCC(=O)N(c1ccccc1)C1CCN(C)CC1,N-Methyl-butyrylfentanyl,fentanyl
+O=C(c1ccco1)N(c1ccccc1)C1CCN(CCc2ccc(O)cc2)CC1,4'-Hydroxy-furanylfentanyl,fentanyl
+CCC(=O)OCC1(N(C(=O)CC)c2ccccc2)CCN(CCN2C(=O)c3ccccc3C2=O)CC1,"{1-[2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)ethyl]-4-[phenyl(propanoyl)amino]piperidin-4-yl}methyl propanoate",fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)C(C)C1,2-Methylfentanyl,fentanyl
+CC(=O)N(c1ccccc1)C1CCN(CCCc2ccccc2)CC1,Phenylpropyl-acetyl-norfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCOc2ccccc2)CC1,POE-F,fentanyl
+O=C(c1ccco1)N(c1ccccc1)C1CCNCC1,Furanyl-norfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(Cc2ccccc2)CC1,R-4129,fentanyl
+Cc1ccc(NC2CCN(CCc3ccccc3)CC2)cc1,p-Me-4-ANPP,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)O)CCN(CCc2ccccc2)C(O)C1,2-Hydroxycarfentanil acid,fentanyl
+COCC(=O)N(c1ccc(F)cc1)C1CCN(CCc2ccccc2)CC1,p-Fluoro-methoxyacetylfentanyl,fentanyl
+Cc1cccc(N(C(=O)c2ccco2)C2CCN(CCc3ccccc3)CC2)c1,m-Methyl-furanylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2F)CC1,2″-Fluorofentanyl,fentanyl
+COc1ccccc1N(C(=O)c1ccco1)C1CCN(CCc2ccccc2)CC1,o-Methoxy-furanylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CC(O)c2ccc(F)cc2)CC1,N-{1-[2-(4-Fluorophenyl)-2-hydroxyethyl]piperidin-4-yl}-N-phenylpropanamide,fentanyl
+CC(=O)N(c1ccccc1)C1CCN(CCc2cccs2)CC1,Acetylthiofentanyl,fentanyl
+COCC(=O)N(c1cccc(F)c1)C1CCN(CCc2ccccc2)CC1,m-Fluoro-methoxyacetylfentanyl,fentanyl
+COc1cc(CCN2CCC(N(C(=O)C(=O)CCC(=O)O)c3ccccc3)CC2)ccc1O,N-Phenyl-N-[1-(2-phenylethyl)piperidin-4-yl]-2-oxo-4-carboxybutanamide,fentanyl
+CCC(=O)N(c1ccccc1)C1(COC)CCN(CCn2nnn(CC)c2=O)CC1,R-39209,fentanyl
+CCC(=O)N(c1ccc(Cl)cc1)C1CCN(CCc2ccccc2)CC1,p-CF,fentanyl
+CCC(=O)N(c1ccccc1)C1CCNCC1,Norfentanyl,fentanyl
+COC(=O)CCN1CCC(Nc2ccccc2)(C(=O)OC)CC1,Despropionyl remifentanil,fentanyl
+CCC(=O)N(c1ccccc1)C1(c2nc(C)cs2)CCN(CCc2ccccc2)CC1,"N-[4-(4-Methyl-1,3-thiazol-2-yl)-1-(2-phenylethyl)piperidin-4-yl]-N-phenylpropanamide",fentanyl
+CCC(=O)N(c1ccccc1)C1(c2ccccc2)CCN(CCc2ccccc2)CC1,4-Phenylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1C)C1CCN(CCc2ccccc2)CC1,o-Methylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCCCCF)CC1,Fluoropentyl-norcarfentanil,fentanyl
+O=C(c1ccco1)N(c1ccccc1)C1CCN(Cc2ccccc2)CC1,Furanylbenzylfentanyl,fentanyl
+CCC(=O)N(c1ccc2ncoc2c1)C1CCN(CCc2ccccc2)CC1,N-Benzoxazolyl-fentanyl,fentanyl
+COCC(=O)N(c1cccc(C)c1)C1CCN(CCc2ccccc2)CC1,m-Methyl-methoxyacetylfentanyl,fentanyl
+CC(C)C(=O)OCC1(N(C(=O)C(C)C)c2ccccc2)CCN(CCN2C(=O)c3ccccc3C2=O)CC1,"{1-[2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)ethyl]-4-[(2-methylpropanoyl)(phenyl)amino]piperidin-4-yl}methyl 2-methylpropanoate",fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCc2ccccn2)CC1,Methyl 4-[phenyl(propanoyl)amino]-1-[2-(pyridin-2-yl)ethyl]piperidine-4-carboxylate,fentanyl
+c1ccc(CCN2CCC(Nc3ccccc3)CC2)cc1,ANPP,fentanyl
+CCCc1cc(OC)c(CCN2CCC(N(C(=O)CC)c3ccccc3)CC2)cc1OC,N-(2C-P) fentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CC(O)c2ccncc2)CC1C,N-{1-[2-Hydroxy-2-(pyridin-4-yl)ethyl]-3-methylpiperidin-4-yl}-N-phenylpropanamide,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CC(=O)c2cccc(F)c2)CC1,N-{1-[2-(3-Fluorophenyl)-2-oxoethyl]piperidin-4-yl}-N-phenylpropanamide,fentanyl
+CCCCc1cc(OC)c(CC(C)N2CCC(N(C(=O)CC)c3ccccc3)CC2)cc1OC,N-(DOBU) fentanyl,fentanyl
+COCC(=O)N(c1ccccc1)C1C(C)CN(CCn2cccn2)CC1C,"N-{3,5-Dimethyl-1-[2-(1H-pyrazol-1-yl)ethyl]piperidin-4-yl}-2-methoxy-N-phenylacetamide",fentanyl
+Cc1ccc(N(C(=O)C2CCCO2)C2CCN(CCc3ccccc3)CC2)cc1,p-Methyl-tetrahydrofuranylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2cc(OC)cc(OC)c2)CC1,"3″,5″-DiMeO-F",fentanyl
+CC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)CC1,Acetylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccc(F)cc2)CC1,4″-Fluorofentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1(COC(=O)OC)CCN(CCc2cccs2)CC1,Methyl {4-[phenyl(propanoyl)amino]-1-[2-(thiophen-2-yl)ethyl]piperidin-4-yl}methyl carbonate,fentanyl
+CCC(=O)[N+]([O-])(c1ccccc1)C1(C(=O)OC)CCN(CCc2ccccc2)CC1,Carfentanil N-oxide,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCc2cccc(O)c2)CC1,3″-Hydroxycarfentanil,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(C(C)C(=O)OC(C)C)CC1,Propan-2-yl 2-{4-[phenyl(propanoyl)amino]piperidin-1-yl}propanoate,fentanyl
+O=C(C1CCC1)N(c1ccc(Cl)cc1)C1CCN(CCc2ccccc2)CC1,p-Chloro-cyclobutylfentanyl,fentanyl
+CCC(C)C(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)CC1,alpha'-Methyl-butyrylfentanyl,fentanyl
+Oc1cc2c(cc1O)CNCC2,Norsalsolinol,neurotoxin
+CC(N)Cc1ccc(Cl)cc1,PCA,neurotoxin
+CC1NCCc2cc(O)c(O)cc21,Salsolinol,neurotoxin
+CC(N)Cc1ccc(Br)cc1,PBA,neurotoxin
+NCCc1cc(O)c(O)cc1O,6-OHDA,neurotoxin
+COc1ccc(CCN)cc1OC,DMPEA,peyote alkaloid
+CNCCc1cc(OC)c(OC)c(OC)c1,M-M,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C(C)N(C=O)CC2,N-Formyl-O-methylanhalonidine,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)[N+](C)(C)CC2,Lophotine,peyote alkaloid
+COc1cc(CCN)ccc1O,GEA,peyote alkaloid
+COc1cc(CCNCC(=O)O)cc(OC)c1OC,Mescaloxylic acid,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)N(C=O)CC2,N-Formylanhalonidine,peyote alkaloid
+COc1cc(CCN2C(=O)CC(O)C2=O)cc(OC)c1OC,Mescaline malimide,peyote alkaloid
+CNCCc1ccc(O)c(O)c1,Epinine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)(C(=O)O)NCC2,Peyoruvic acid,peyote alkaloid
+CCN1CCc2cc(OC)c3c(c2C1C)OCO3,Peyophorine,peyote alkaloid
+COc1cc(CCN2C(=O)C=CC2=O)cc(OC)c1OC,Mescaline maleimide,peyote alkaloid
+COc1c(O)cc2c(c1OC)CN(C)CC2,Isoanhalidine,peyote alkaloid
+NCCc1ccc2c(c1)OCO2,MDPEA,peyote alkaloid
+COc1cc2c(c(O)c1OC)C1CCC(=O)N1CC2,Peyoglutam,peyote alkaloid
+COc1cc(CCN)cc(O)c1OC,3-DESMETHYL,peyote alkaloid
+CCOc1ccc(CC(C)N(C)C)c(OCC)c1,"2,4-Diethoxy-N,N-dimethylamphetamine",peyote alkaloid
+CN(C)CCc1ccc2c(c1)OCO2,Lobivine,peyote alkaloid
+COc1c(O)cc2c(c1OC)C(C)N(C)CC2,Isopellotine,peyote alkaloid
+COc1cc(CCNC(C)=O)cc(OC)c1OC,N-Acetylmescaline,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)N(C(C)=O)CC2,N-Acetylanhalonine,peyote alkaloid
+COc1cc(CCN(C)C)cc(O)c1OC,"N,N-Me-3-DESMETHYL",peyote alkaloid
+COc1c(O)cc2c(c1OC)C(C)NCC2,Isoanhalonidine,peyote alkaloid
+COc1cc(CCn2cccc2C(=O)O)cc(OC)c1OC,Peyonine,peyote alkaloid
+COc1cc2c(c(O)c1OC)CN(C(C)=O)CC2,N-Acetylanhalamine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C(=O)O)NCC2,Peyoxylic acid,peyote alkaloid
+COc1cc(CCN(C)C)ccc1O,MM-GEA,peyote alkaloid
+COc1cc2c(c(OC)c1OC)CNCC2,Anhalinine,peyote alkaloid
+CNCCc1ccc(O)cc1,HMePEA,peyote alkaloid
+COc1cc2c(c(O)c1OC)C[N+](C)(C)CC2,Anhalotine,peyote alkaloid
+COc1c(O)cc2c(c1OC)CNCC2,Isoanhalamine,peyote alkaloid
+COc1cc(CCN)cc(O)c1O,"3,4-DESMETHYL",peyote alkaloid
+COc1cc2c(c(O)c1OC)CN(C=O)CC2,N-Formylanhalamine,peyote alkaloid
+COc1cc(CCNC(C)C(=O)O)cc(OC)c1OC,Mescaloruvic acid,peyote alkaloid
+C[N+](C)(C)CCc1ccc(O)cc1,Candicine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)NCC2,Anhalonidine,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C1CCC(=O)N1CC2,Mescalotam,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)NCC2,Anhalonine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)N(C)CC2,Pellotine,peyote alkaloid
+COc1cc2c(c(O)c1OC)CN(C)CC2,Anhalidine,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)N(C)CC2,Lophophorine,peyote alkaloid
+COc1cc(CCN)cc2c1OCO2,LOPHOPHINE,peyote alkaloid
+COc1cc(CCNC(C)=O)cc(O)c1OC,N-Acetyl-3-demethymescaline,peyote alkaloid
+COc1cc(CCN)cc(OC)c1OC,M,peyote alkaloid
+COc1cc(CCN2C(=O)CCC2=O)cc(OC)c1OC,Mescaline succinimide,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)[N+](C)(C)CC2,Peyotine,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C(C)NCC2,O-Methylanhalonidine,peyote alkaloid
+COc1cc2c(c(O)c1OC)CNCC2,Anhalamine,peyote alkaloid
+COc1cc(CCN2C(=O)CC(O)(CC(=O)O)C2=O)cc(OC)c1OC,Mescaline citrimide,peyote alkaloid
+CN(C)CCc1ccc(O)cc1,Hordenine,peyote alkaloid
+COc1cc(CCNC=O)cc(OC)c1OC,N-Formylmescaline,peyote alkaloid
+CNCCc1cc(O)c(OC)c(OC)c1,N-Me-3-DESMETHYL,peyote alkaloid
+COc1cc(CCn2c(C=O)ccc2CO)cc(OC)c1OC,Peyoglunal,peyote alkaloid
+COc1cc(CCNC=O)cc(O)c1OC,N-Formyl-3-demethylmescaline,peyote alkaloid
+NCCc1ccc(O)c(O)c1,Dopamine,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)N(C=O)CC2,N-Formylanhalonine,peyote alkaloid
+CNCCc1ccc(O)c(OC)c1,N-Me-GEA,peyote alkaloid
+COc1cc2c(c(OC)c1OC)CN(C=O)CC2,N-Formylanhalinine,peyote alkaloid
+COc1cc(CCN2C(=O)C3CC(=O)OC3C2=O)cc(OC)c1OC,Mescaline isocitrimide lactone,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C(C)N(C)CC2,O-Methylpellotine,peyote alkaloid
+CCC(C)N(CCc1c[nH]c2ccc(F)cc12)C(C)C,5-Fluoro-IPSBT,tryptamine
+CC(C)N(CCc1c[nH]c2ccc(F)cc12)CC1CC1,5-Fluoro-N-cyclopropylmethyl-NIPT,tryptamine
+CCCCCCN(C)CCc1c[nH]c2ccc(OC)cc12,N-[2-(5-Methoxy-1H-indol-3-yl)ethyl]-N-methylhexan-1-amine,tryptamine
+CN(C)CCc1c[nH]c2c(Br)ccc(O)c12,7-Bromopsilocin,tryptamine
+CCCCN(CCCC)CCc1c[nH]c2cc(OC)c(OC)cc12,"5,6-MeO-DBT",tryptamine
+CN(C/C=C/c1ccccc1)CCc1c[nH]c2ccccc12,N-Phenylpropenyl-NMT,tryptamine
+COc1cc2[nH]cc(CCN3CCCC3)c2cc1OC,"5,6-MeO-pyr-T",tryptamine
+Cn1cc(CCN)c2c3c(ccc21)OCCC3,"CP 132,484",tryptamine
+C=CCN(CCc1c(C)[nH]c2ccc(OC)cc12)C1CCCCC1,2-Methyl-5-methoxy-N-allyl-N-cyclohexyltryptamine,tryptamine
+Cc1cc(OP(=O)(O)O)c2c(CCN)c[nH]c2c1,6-Methyl-4-PO-T,tryptamine
+CCCCCN(CCCCC)CCc1c[nH]c2ccccc12,"N,N-Dipentyltryptamine",tryptamine
+COc1ccc2[nH]cc(CC3CCCN3C)c2c1,CP-108509,tryptamine
+COc1ccc(OC)c(CCCN(C)CCc2c[nH]c3ccc(OC)cc23)c1,"3-(2,5-Dimethoxyphenyl)-N-[2-(5-methoxy-1H-indol-3-yl)ethyl]-N-methylpropan-1-amine",tryptamine
+C=CCN(CCc1c[nH]c2ccc(F)cc12)C(C)C,5-Fluoro-ALIPT,tryptamine
+NCCc1c[nH]c2ccc(OC3CCCC3)cc12,5-Cyclopentyloxy-T,tryptamine
+COc1ccc(OC)c(CCN(C)CCc2c[nH]c3ccc(OC)cc23)c1,"N-(2,5-Dimethoxyphenylethyl)-5-MeO-NMT",tryptamine
+Oc1cccc2[nH]cc(CCN3CCCCC3)c12,4-HO-pip-T,tryptamine
+CN(C)CCc1c[nH]c2cccc(OCc3ccccc3)c12,4-Benzyloxy-DMT,tryptamine
+CN(C)CCc1c[nH]c2cccc(OC(=O)SCc3ccccc3)c12,S-Benzyl O-{3-[2-(dimethylamino)ethyl]-1H-indol-4-yl} carbonothioate,tryptamine
+COc1cc(OC)c2[nH]cc(CCNCc3ccccc3)c2c1,"5,7-MeO-T-NB",tryptamine
+CNC(C)Cc1c[nH]c2ccc(OCc3ccccc3)cc12,5-Benzyloxy-alpha-methyl-NMT,tryptamine
+CC(C)(C)CCCCCCCOc1ccc2[nH]cc(CCN)c2c1,"5-(8,8-Dimethylnonyloxy)-T",tryptamine
+CCCOc1ccc2[nH]cc(CCN(CC)CC)c2c1,5-PrO-DET,tryptamine
+COc1ccc2[nH]cc(CCN(C)CCCc3cc(OC)c(OC)c(OC)c3)c2c1,"N-(3,4,5-Trimethoxyphenylpropyl)-5-MeO-NMT",tryptamine
+c1ccc2c(CCN3CCN(CCc4c[nH]c5ccccc45)CC3)c[nH]c2c1,"3,3'-[Piperazine-1,4-diyldi(ethane-2,1-diyl)]di(1H-indole)",tryptamine
+CC(C)N(CCc1c[nH]c2cccc(OCc3ccccc3)c12)C(C)C,4-Benzyloxy-DIPT,tryptamine
+CC(C)N(CCc1c[nH]c2ccc(F)cc12)C1CCC1,5-Fluoro-IPCBT,tryptamine
+CCOc1ccc(CNCCc2c[nH]c3ccccc23)cc1,T-NB4OEt,tryptamine
+O=C(O)CCNCCc1c[nH]c2cc([N+](=O)[O-])ccc12,6-Nitro-N-(2-carboxyethyl)tryptamine,tryptamine
+CN1CCN(CCc2c[nH]c3cccc(O)c23)CC1,4'-Methyl-4-HO-piperazine-T,tryptamine
+CN(CCC#N)CCc1c[nH]c2ccccc12,N-Cyanoethyl-NMT,tryptamine
+COc1cc(F)c2[nH]cc(CCN(C)C)c2c1,7-Fluoro-5-MeO-DMT,tryptamine
+CN(CCc1c[nH]c2ccc(F)cc12)C1CCC1,5-Fluoro-MCBT,tryptamine
+CNCCc1c[nH]c2c(OC)cccc12,7-MeO-NMT,tryptamine
+Cc1c(O)c(O)cc2c(CCN)c[nH]c12,"7-Methyl-5,6-HO-T",tryptamine
+Oc1ccc(CNCCc2c[nH]c3ccccc23)cc1,T-NB4OH,tryptamine
+COc1c(Br)cc(Br)cc1CNCCc1c[nH]c2ccccc12,T-NBOMe35Br,tryptamine
+CCn1cc(CCN(C)C)c2c(OCc3ccccc3)cc(F)cc21,1-Ethyl-6-fluoro-4-benzyloxy-DMT,tryptamine
+CN(C)CCc1c[nH]c2cccc(OC(=O)c3cccc(C(=O)Oc4cccc5[nH]cc(CCN(C)C)c45)c3)c12,"Bis{3-[2-(dimethylamino)ethyl]-1H-indol-4-yl} benzene-1,3-dicarboxylate",tryptamine
+CCN(CC)C(=O)CCN(C)CCc1c[nH]c2ccc(OC)cc12,"N,N-Diethyl-N3-[2-(5-methoxy-1H-indol-3-yl)ethyl]-N3-methyl-beta-alaninamide",tryptamine
+COc1ccc(CNCCc2c[nH]c3c(Br)cc(OC)cc23)cc1,7-Bromo-5-MeO-T-NB4OMe,tryptamine
+CCCN(C)CCc1c[nH]c2ccc(F)cc12,5-Fluoro-MPT,tryptamine
+CN(CCc1c[nH]c2cc3c(cc12)OCO3)Cc1ccccc1,"5,6-MDO-NMT-NB",tryptamine
+Cc1c(OC(=O)CCC(=O)O)ccc2[nH]cc(CCN(C)C)c12,Psilocin-4-methyl-5-succinate,tryptamine
+COc1ccc2[nH]cc(CC(C)N)c2c1Br,4-Bromo-5-MeO-AMT,tryptamine
+CN(C)CC(=O)c1c[nH]c2cc(F)ccc12,beta-Oxo-6-fluoro-DMT,tryptamine
+COc1ccc2[nH]cc(C(O)CN(C)C)c2c1,beta-HO-5-MeO-DMT,tryptamine
+COc1cccc2c(CCN)c[nH]c12,7-MeO-T,tryptamine
+COc1cc(I)c2[nH]cc(CCN(C)C)c2c1,7-Iodo-5-MeO-DMT,tryptamine
+C[C@@H]1CC[C@H](C)N1CCc1c[nH]c2ccccc12,"(R,S)-Dimethyl-pyr-T",tryptamine
+COc1ccc(Sn2cc(CCN(C)C)c3c(OC)cccc32)cc1,1-(4-Methoxyphenylthio)-4-MeO-DMT,tryptamine
+CCCN(CCc1c[nH]c2c(F)cccc12)C1CCC1,7-Fluoro-PCBT,tryptamine
+CC(=O)NCCc1c[nH]c2cc(F)cc(NC(C)=O)c12,6-Fluoro-4-acetamido-N-acetyltryptamine,tryptamine
+CCCCS(=O)(=O)n1cc(CCN(C)C)c2c(OC)cccc21,1-Butylsulfonyl-4-MeO-DMT,tryptamine
+CCCN(CCC)CCc1c[nH]c2ccc(OC(C)=O)cc12,5-AcO-DPT,tryptamine
+CN(C)CCc1cn(C)c2ccccc12,1-TMT,tryptamine
+CCN(CCc1c[nH]c2cc(F)c(F)cc12)C(C)C,"5,6-Difluoro-EIPT",tryptamine
+CN(C)CC(=O)c1c[nH]c2c(F)cc(F)cc12,"beta-Oxo-5,7-difluoro-DMT",tryptamine
+CN(C)CCc1cn(C)c2cc(F)cc(OCc3ccccc3)c12,1-Methyl-6-fluoro-4-benzyloxy-DMT,tryptamine
+CCCc1[nH]c2ccc(OC)cc2c1CCN(C)C,2-Propyl-5-MeO-DMT,tryptamine
+CC(=O)Oc2cccc1[nH]cc(CCN(C)C)c12,4-AcO-DMT,tryptamine

src/data/sample_smiles.csv

@@ -0,0 +1,13 @@
+smiles,name,category
+CCO,Ethanol,Alcohol
+CC(=O)O,Acetic acid,Carboxylic acid
+c1ccccc1,Benzene,Aromatic hydrocarbon
+CC(C)O,Isopropanol,Alcohol
+c1ccc(cc1)O,Phenol,Aromatic alcohol
+CC(=O)OC1=CC=CC=C1C(=O)O,Aspirin,Carboxylic acid ester
+CN1C=NC2=C1C(=O)N(C(=O)N2C)C,Caffeine,Alkaloid
+CCN(CC)CC,Triethylamine,Amine
+c1ccc2c(c1)cccn2,Quinoline,Aromatic heterocycle
+c1ccc(cc1)N,Aniline,Aromatic amine
+O=P(O)(O)OP(=O)(O)OP(=O)(O)OC[C@H]3O[C@@H](n2cnc1c(ncnc12)N)[C@H](O)[C@@H]3O,ATP,Nucleotide
+CN(C)CCC1=CNC2=C1C(=CC=C2)OP(=O)(O)O,Psilocybin,Tryptamine

src/main.js

@@ -0,0 +1,322 @@
+/**
+ * @fileoverview JSME (JavaScript Molecule Editor) integration with Gradio interface
+ * Handles bidirectional synchronization between JSME applet and Gradio textbox
+ * @author Manny Cortes ('[email protected]')
+ * @version 0.2.0
+ */
+
+// ============================================================================
+// GLOBAL VARIABLES
+// ============================================================================
+
+/** @type {Object|null} The JSME applet instance */
+let jsmeApplet = null;
+
+/** @type {string} Last known value of the textbox to prevent infinite loops */
+let lastTextboxValue = "";
+
+// ============================================================================
+// CONSTANTS
+// ============================================================================
+
+/** @const {string} Default SMILES for initial molecule (ethanol) */
+const DEFAULT_SMILES = "CCO";
+
+/** @const {string} Container height for JSME applet */
+const CONTAINER_HEIGHT = "450px";
+
+/** @const {string} CSS selector for the Gradio SMILES input element */
+const SMILES_INPUT_SELECTOR = "#smiles_input textarea, #smiles_input input";
+
+/** @const {number} Delay for paste event handling (ms) */
+const PASTE_DELAY = 50;
+
+/** @const {number} Delay for initialization retry (ms) */
+const INIT_RETRY_DELAY = 2000;
+
+/** @const {string[]} Events to trigger for Gradio change detection */
+const GRADIO_CHANGE_EVENTS = ["input", "change", "keyup"];
+
+// ============================================================================
+// CORE INITIALIZATION
+// ============================================================================
+
+/**
+ * Initializes the JSME applet after the library has been loaded
+ * Sets up the molecular editor with default options and callbacks
+ * @throws {Error} When JSME initialization fails
+ */
+function initializeJSME() {
+  try {
+    console.log("Initializing JSME...");
+    // https://github.com/jsme-editor/jsme-editor.github.io
+    // https://jsme-editor.github.io/dist/api_javadoc/index.html
+    // http://wiki.jmol.org/index.php/Jmol_JavaScript_Object/JME/Options
+    jsmeApplet = new JSApplet.JSME(
+      "jsme_container",
+      getJsmeContainerWidthPx(),
+      CONTAINER_HEIGHT,
+      {
+        options:
+          "rButton,zoom,zoomgui,newLook,star,multipart,polarnitro,NOexportInChI,NOexportInChIkey,NOsearchInChIkey,NOexportSVG,NOpaste",
+      }
+    );
+
+    jsmeApplet.setCallBack("AfterStructureModified", handleJSMEStructureChange);
+    jsmeApplet.setMenuScale(getJsmeGuiScale());
+    jsmeApplet.setUserInterfaceBackgroundColor("#adadad");
+
+    // Set initial molecule and sync state
+    jsmeApplet.readGenericMolecularInput(DEFAULT_SMILES);
+    lastTextboxValue = DEFAULT_SMILES;
+
+    setupTextboxEventListeners();
+    window.addEventListener("resize", handleResize);
+
+    console.log("JSME initialized successfully");
+  } catch (error) {
+    console.error("Error initializing JSME:", error);
+    throw error;
+  }
+}
+
+/**
+ * Handles structure changes in the JSME applet
+ * Converts the structure to SMILES and updates the Gradio textbox
+ * @param {Event} event - The JSME structure modification event
+ */
+function handleJSMEStructureChange(event) {
+  try {
+    const smiles = jsmeApplet.smiles();
+    updateGradioTextbox(smiles);
+  } catch (error) {
+    console.error("Error getting SMILES from JSME:", error);
+  }
+}
+
+/**
+ * Calculates the appropriate GUI scale for the JSME applet based on container width
+ * Uses breakpoints to determine optimal scaling for different screen sizes
+ * @returns {number} The scale factor for the JSME GUI (0.88 to 2.0)
+ */
+function getJsmeGuiScale() {
+  const width = getJsmeContainerWidthNumber();
+  let menuScale;
+  if (width > 460) {
+    menuScale = 1.3;
+  } else if (width > 420) {
+    menuScale = 1.1;
+  } else if (width > 370) {
+    menuScale = 1.05;
+  } else if (width > 300) {
+    menuScale = 0.88;
+  } else {
+    menuScale = 2;
+  }
+  return menuScale;
+}
+
+/**
+ * Gets the JSME container width as a CSS-compatible string value
+ * Returns either a pixel value or percentage based on available width
+ * @returns {string} Width as "100%" or "{width}px" format
+ */
+function getJsmeContainerWidthPx() {
+  const parentWidth = getJsmeContainerWidthNumber();
+  if (parentWidth == null || parentWidth <= 0) {
+    return "100%";
+  }
+  return `${parentWidth}px`;
+}
+
+/**
+ * Gets the numeric width of the JSME container's parent element
+ * Used for responsive scaling calculations
+ * @returns {number|null} Width in pixels, or null if container not found
+ */
+function getJsmeContainerWidthNumber() {
+  const container = document.getElementById("jsme_container");
+  if (!container) {
+    return null;
+  }
+  return container.parentNode.offsetWidth;
+}
+
+// ============================================================================
+// GRADIO INTEGRATION
+// ============================================================================
+
+/**
+ * Updates the Gradio textbox with a SMILES string
+ * Triggers appropriate events to ensure Gradio detects the change
+ * @param {string} smiles - The SMILES string to set in the textbox
+ */
+function updateGradioTextbox(smiles) {
+  try {
+    const textbox = document.querySelector(SMILES_INPUT_SELECTOR);
+
+    if (!textbox || textbox.value === smiles) {
+      return;
+    }
+
+    textbox.value = smiles;
+    lastTextboxValue = smiles;
+
+    // Trigger events to ensure Gradio detects the change
+    GRADIO_CHANGE_EVENTS.forEach((eventType) => {
+      const event = new Event(eventType, {
+        bubbles: true,
+        cancelable: true,
+      });
+      textbox.dispatchEvent(event);
+    });
+  } catch (error) {
+    console.error("Error updating Gradio textbox:", error);
+  }
+}
+
+// ============================================================================
+// JSME UPDATE FUNCTIONS
+// ============================================================================
+
+/**
+ * Updates the JSME applet with a SMILES string from the textbox
+ * @param {string} smiles - The SMILES string to display in JSME
+ */
+function updateJSMEFromTextbox(smiles) {
+  if (!jsmeApplet) {
+    return;
+  }
+
+  try {
+    if (smiles && smiles.trim() !== "") {
+      jsmeApplet.readGenericMolecularInput(smiles.trim());
+    } else {
+      jsmeApplet.reset();
+    }
+    lastTextboxValue = smiles;
+  } catch (error) {
+    console.error("Error updating JSME from textbox:", error);
+  }
+}
+
+// ============================================================================
+// UI MONITORING
+// ============================================================================
+
+/**
+ * Finds the textbox element and sets up event listeners
+ */
+function setupTextboxEventListeners() {
+  const textbox = document.querySelector(SMILES_INPUT_SELECTOR);
+  if (!textbox) {
+    return;
+  }
+
+  textbox.addEventListener("input", handleTextboxChange);
+  textbox.addEventListener("change", handleTextboxChange);
+  textbox.addEventListener("paste", handleTextboxPaste);
+  textbox.addEventListener("keyup", handleTextboxChange);
+}
+
+/**
+ * Handles textbox change events
+ * @param {Event} event - The change event
+ */
+function handleTextboxChange(event) {
+  if (event.target.value !== lastTextboxValue) {
+    updateJSMEFromTextbox(event.target.value);
+  }
+}
+
+/**
+ * Handles textbox paste events with a delay to ensure content is available
+ * @param {Event} event - The paste event
+ */
+function handleTextboxPaste(event) {
+  setTimeout(() => {
+    updateJSMEFromTextbox(event.target.value);
+  }, PASTE_DELAY);
+}
+
+/**
+ * Handles window resize events and updates JSME applet width
+ */
+function handleResize() {
+  if (!jsmeApplet) {
+    return;
+  }
+
+  try {
+    jsmeApplet.setMenuScale(getJsmeGuiScale());
+    jsmeApplet.setWidth(getJsmeContainerWidthPx());
+  } catch (error) {
+    console.error("Error resizing JSME applet:", error);
+  }
+}
+
+// ============================================================================
+// PUBLIC API
+// ============================================================================
+
+/**
+ * Sets a SMILES string in both JSME and Gradio textbox
+ * @param {string} smiles - The SMILES string to set
+ * @returns {string} The SMILES string that was set
+ * @public
+ */
+window.setJSMESmiles = function (smiles) {
+  if (jsmeApplet) {
+    updateJSMEFromTextbox(smiles);
+  }
+
+  updateGradioTextbox(smiles);
+  return smiles;
+};
+
+/**
+ * Clears both JSME and Gradio textbox
+ * @returns {Array} Array containing cleared state for Gradio components
+ * @public
+ */
+window.clearJSME = function () {
+  if (jsmeApplet) {
+    jsmeApplet.reset();
+  }
+
+  updateGradioTextbox("");
+  return ["", [], [], "Cleared - Draw a new molecule or enter SMILES"];
+};
+
+// ============================================================================
+// INITIALIZATION LOGIC
+// ============================================================================
+
+/**
+ * Checks if JSME library is loaded and initializes JSME applet
+ * Retries until the library becomes available
+ */
+function initializeWhenReady() {
+  if (typeof JSApplet !== "undefined" && JSApplet.JSME) {
+    console.log("JSME library loaded, initializing...");
+    initializeJSME();
+  } else {
+    console.log("JSME library not ready, retrying...");
+    setTimeout(initializeWhenReady, INIT_RETRY_DELAY);
+  }
+}
+
+/**
+ * Starts the initialization process based on document ready state
+ */
+function startInitialization() {
+  if (document.readyState === "loading") {
+    document.addEventListener("DOMContentLoaded", () => {
+      setTimeout(initializeWhenReady, INIT_RETRY_DELAY);
+    });
+  } else {
+    setTimeout(initializeWhenReady, INIT_RETRY_DELAY);
+  }
+}
+
+startInitialization();

src/main.min.js

@@ -0,0 +1 @@
+function initializeJSME(){try{(jsmeApplet=new JSApplet.JSME("jsme_container",getJsmeContainerWidthPx(),CONTAINER_HEIGHT,{options:"rButton,zoom,zoomgui,newLook,star,multipart,polarnitro,NOexportInChI,NOexportInChIkey,NOsearchInChIkey,NOexportSVG,NOpaste"})).setCallBack("AfterStructureModified",handleJSMEStructureChange),jsmeApplet.setMenuScale(getJsmeGuiScale()),jsmeApplet.setUserInterfaceBackgroundColor("#adadad"),jsmeApplet.readGenericMolecularInput(DEFAULT_SMILES),lastTextboxValue=DEFAULT_SMILES,setupTextboxEventListeners(),window.addEventListener("resize",handleResize)}catch(e){throw e}}function handleJSMEStructureChange(){try{updateGradioTextbox(jsmeApplet.smiles())}catch(e){}}function getJsmeGuiScale(){const e=getJsmeContainerWidthNumber();return e>460?1.3:e>420?1.1:e>370?1.05:e>300?.88:2}function getJsmeContainerWidthPx(){const e=getJsmeContainerWidthNumber();return null==e||e<=0?"100%":`${e}px`}function getJsmeContainerWidthNumber(){const e=document.getElementById("jsme_container");return e?e.parentNode.offsetWidth:null}function updateGradioTextbox(e){try{const t=document.querySelector(SMILES_INPUT_SELECTOR);if(!t||t.value===e)return;t.value=e,lastTextboxValue=e,GRADIO_CHANGE_EVENTS.forEach(e=>{const n=new Event(e,{bubbles:!0,cancelable:!0});t.dispatchEvent(n)})}catch(e){}}function updateJSMEFromTextbox(e){if(jsmeApplet)try{e&&""!==e.trim()?jsmeApplet.readGenericMolecularInput(e.trim()):jsmeApplet.reset(),lastTextboxValue=e}catch(e){}}function setupTextboxEventListeners(){const e=document.querySelector(SMILES_INPUT_SELECTOR);e&&(e.addEventListener("input",handleTextboxChange),e.addEventListener("change",handleTextboxChange),e.addEventListener("paste",handleTextboxPaste),e.addEventListener("keyup",handleTextboxChange))}function handleTextboxChange(e){e.target.value!==lastTextboxValue&&updateJSMEFromTextbox(e.target.value)}function handleTextboxPaste(e){setTimeout(()=>{updateJSMEFromTextbox(e.target.value)},PASTE_DELAY)}function handleResize(){if(jsmeApplet)try{jsmeApplet.setMenuScale(getJsmeGuiScale()),jsmeApplet.setWidth(getJsmeContainerWidthPx())}catch(e){}}function initializeWhenReady(){"undefined"!=typeof JSApplet&&JSApplet.JSME?initializeJSME():setTimeout(initializeWhenReady,INIT_RETRY_DELAY)}function startInitialization(){"loading"===document.readyState?document.addEventListener("DOMContentLoaded",()=>{setTimeout(initializeWhenReady,INIT_RETRY_DELAY)}):setTimeout(initializeWhenReady,INIT_RETRY_DELAY)}let jsmeApplet=null,lastTextboxValue="";const DEFAULT_SMILES="CCO",CONTAINER_HEIGHT="450px",SMILES_INPUT_SELECTOR="#smiles_input textarea, #smiles_input input",PASTE_DELAY=50,INIT_RETRY_DELAY=2e3,GRADIO_CHANGE_EVENTS=["input","change","keyup"];window.setJSMESmiles=function(e){return jsmeApplet&&updateJSMEFromTextbox(e),updateGradioTextbox(e),e},window.clearJSME=function(){return jsmeApplet&&jsmeApplet.reset(),updateGradioTextbox(""),["",[],[],"Cleared - Draw a new molecule or enter SMILES"]},startInitialization();

src/service.py

@@ -0,0 +1,213 @@
+import logging
+import os
+from typing import TypedDict
+
+import numpy as np
+import pandas as pd
+import redis
+from chem_mrl.molecular_fingerprinter import MorganFingerprinter
+from dotenv import load_dotenv
+from redis.commands.search.field import TextField, VectorField
+from redis.commands.search.indexDefinition import IndexDefinition, IndexType
+from redis.commands.search.query import Query
+from sentence_transformers import SentenceTransformer
+
+from constants import (
+    EMBEDDING_DIMENSION,
+    HNSW_K,
+    HNSW_PARAMETERS,
+    MODEL_NAME,
+    SUPPORTED_EMBEDDING_DIMENSIONS,
+)
+from data import DATASET_SMILES, ISOMER_DESIGN_SUBSET
+
+
+def setup_logger(clear_handler=False):
+    if clear_handler:
+        for handler in logging.root.handlers[:]:
+            logging.root.removeHandler(handler)  # issue with sentence-transformer's logging handler
+    logging.basicConfig(format="%(asctime)s - %(message)s", datefmt="%Y-%m-%d %H:%M:%S", level=logging.INFO)
+    logger = logging.getLogger(__name__)
+    return logger
+
+
+load_dotenv("../.env")
+logger = setup_logger(clear_handler=True)
+
+
+class SimilarMolecule(TypedDict):
+    smiles: str
+    name: str
+    category: str
+    score: float
+
+
+class MolecularEmbeddingService:
+    def __init__(self):
+        self.model_name = MODEL_NAME
+        self.index_name = "molecule_embeddings"
+        self.model_embed_dim = EMBEDDING_DIMENSION
+
+        self.model = self._initialize_model()
+        self.redis_client = self._initialize_redis()
+        self._initialize_datastore()
+
+    def _initialize_model(self):
+        """Initialize the Hugging Face transformers model"""
+        try:
+            model = SentenceTransformer(self.model_name)  # type: ignore
+            model.eval()
+            return model
+        except Exception as e:
+            logger.error(f"Failed to load model: {e}")
+            raise
+
+    def _initialize_redis(self):
+        """Initialize Redis connection"""
+        try:
+            redis_host = os.getenv("REDIS_HOST", "localhost")
+            redis_port = int(os.getenv("REDIS_PORT", 6379))
+            redis_password = os.getenv("REDIS_PASSWORD", None)
+            logger.info(
+                f"Connecting to Redis at {redis_host}:{redis_port} with password: {'***' if redis_password else 'None'}"
+            )
+            redis_client = redis.Redis(
+                host=redis_host,
+                port=redis_port,
+                password=redis_password,
+                decode_responses=True,
+            )
+            redis_client.ping()
+            return redis_client
+        except Exception as e:
+            logger.error(f"Failed to connect to Redis: {e}")
+            raise
+
+    def _initialize_datastore(self):
+        self.__create_hnsw_index()
+        self.__populate_sample_data(DATASET_SMILES)
+        self.__populate_sample_data(ISOMER_DESIGN_SUBSET)
+
+    def __create_hnsw_index(self):
+        """Create HNSW index for molecular embeddings"""
+        try:
+            self.redis_client.ft(self.index_name).info()
+            logger.info(f"Index {self.index_name} already exists")
+            return
+        except redis.exceptions.ResponseError:
+            pass
+
+        try:
+            schema: list[TextField | VectorField] = [
+                VectorField(
+                    self.embedding_field_name(dim),
+                    "HNSW",
+                    {
+                        **HNSW_PARAMETERS,
+                        "DIM": dim,
+                    },
+                )
+                for dim in SUPPORTED_EMBEDDING_DIMENSIONS
+            ]
+            schema.insert(0, TextField("smiles"))
+
+            self.redis_client.ft(self.index_name).create_index(
+                schema,
+                definition=IndexDefinition(prefix=[self.molecule_index_prefix("")], index_type=IndexType.HASH),
+            )
+
+            logger.info(f"Created HNSW index: {self.index_name}")
+
+        except Exception as e:
+            logger.error(f"Failed to create HNSW index: {e}")
+            raise
+
+    def __populate_sample_data(self, df: pd.DataFrame):
+        """Populate Redis with sample molecular data"""
+        logger.info("Populating Redis with sample molecular data...")
+        for _, row in df.iterrows():
+            try:
+                key = self.molecule_index_prefix(row["smiles"])
+                if self.redis_client.exists(key):
+                    continue
+
+                mapping: dict[str, bytes | str] = {
+                    self.embedding_field_name(embed_dim): self.get_molecular_embedding(
+                        row["smiles"], embed_dim
+                    ).tobytes()
+                    for embed_dim in SUPPORTED_EMBEDDING_DIMENSIONS
+                }
+                mapping = {**mapping, **row.to_dict()}
+
+                self.redis_client.hset(
+                    key,
+                    mapping=mapping,
+                )
+
+            except Exception as e:
+                logger.error(f"Failed to process molecule {row}: {e}")
+                continue
+
+        logger.info(f"Populated {len(df)} sample molecules")
+
+    def get_molecular_embedding(self, smiles: str, embed_dim: int) -> np.ndarray:
+        """Generate molecular embedding using ChemMRL"""
+        try:
+            if embed_dim <= 0:
+                raise ValueError("embed_dim must be positive")
+
+            # Preprocess smiles similarly as training data for optimal performance
+            smiles = MorganFingerprinter.canonicalize_smiles(smiles) or smiles
+
+            embedding: np.ndarray = self.model.encode(
+                [smiles],
+                show_progress_bar=False,
+                convert_to_numpy=True,
+            )[0]
+
+            return self._truncate_and_normalize_embedding(embedding, embed_dim)
+
+        except Exception as e:
+            logger.error(f"Failed to generate embedding for {smiles}: {e}")
+            raise
+
+    def _truncate_and_normalize_embedding(self, embedding: np.ndarray, embed_dim: int) -> np.ndarray:
+        """Truncate and normalize embedding"""
+        if embed_dim < len(embedding):
+            embedding = embedding[:embed_dim]
+        norms = np.linalg.norm(embedding, ord=2, keepdims=True)
+        return embedding / np.where(norms == 0, 1, norms)
+
+    def find_similar_molecules(
+        self, query_embedding: np.ndarray, embed_dim: int, k: int = HNSW_K
+    ) -> list[SimilarMolecule]:
+        """Find k most similar molecules using HNSW"""
+        try:
+            query_vector = query_embedding.astype(np.float32).tobytes()
+            query = (
+                Query(f"*=>[KNN {k} @{self.embedding_field_name(embed_dim)} $vec AS score]")
+                .sort_by("score")
+                .return_fields("smiles", "name", "category", "score")
+                .dialect(2)
+            )
+
+            results = self.redis_client.ft(self.index_name).search(query, query_params={"vec": query_vector})
+
+            neighbors: list[SimilarMolecule] = [
+                {"smiles": doc.smiles, "name": doc.name, "category": doc.category, "score": float(doc.score)}
+                for doc in results.docs
+            ]
+
+            return neighbors
+
+        except Exception as e:
+            logger.error(f"Failed to find similar molecules: {e}")
+            return []
+
+    @staticmethod
+    def embedding_field_name(dim: int) -> str:
+        return f"embedding_{dim}"
+
+    @staticmethod
+    def molecule_index_prefix(smiles: str) -> str:
+        return f"mol:{smiles}"

src/src/app.py

@@ -0,0 +1,218 @@
+import gradio as gr
+import numpy as np
+import pandas as pd
+from rdkit import Chem
+from rdkit.Chem import Draw
+from rdkit.Chem.Draw import rdMolDraw2D
+
+from constants import EMBEDDING_DIMENSION, LAUNCH_PARAMETERS, SUPPORTED_EMBEDDING_DIMENSIONS
+from data import SAMPLE_SMILES
+from service import MolecularEmbeddingService, SimilarMolecule, setup_logger
+
+logger = setup_logger()
+
+
+class App:
+    def __init__(self):
+        self.embedding_service = MolecularEmbeddingService()
+        self.demo = self.create_gradio_interface()
+
+    def molecule_similarity_search_pipeline(
+        self, smiles: str, embed_dim: int
+    ) -> tuple[list[float], list[SimilarMolecule], str]:
+        """Complete pipeline: SMILES -> Canonical SMILES -> Embedding -> Similar molecules"""
+        try:
+            if not smiles or smiles.strip() == "":
+                return [], [], "Please provide a valid SMILES string"
+
+            logger.info(f"Running similarity search: {smiles} - ({embed_dim})")
+            embedding = self.embedding_service.get_molecular_embedding(smiles, embed_dim)
+            neighbors = self.embedding_service.find_similar_molecules(embedding, embed_dim)
+
+            return embedding.tolist(), neighbors, "Search completed successfully"
+
+        except Exception as e:
+            error_msg = f"Search failed: {str(e)}"
+            logger.error(error_msg)
+            return [], [], error_msg
+
+    @staticmethod
+    def _truncated_attribute(obj, attr, max_len=45):
+        return f"{obj[attr][:max_len]}{'...' if len(obj[attr]) > max_len else ''}"
+
+    @classmethod
+    def _draw_molecule_grid(cls, similar: list[SimilarMolecule]) -> np.ndarray:
+        mols = [Chem.MolFromSmiles(m["smiles"]) for m in similar]
+        legends = [
+            f"{cls._truncated_attribute(m, 'name')}\n{m['category']}\n"
+            f"{cls._truncated_attribute(m, 'smiles')}\n({m['score']:.2E})"
+            for m in similar
+        ]
+
+        draw_options = rdMolDraw2D.MolDrawOptions()
+        draw_options.legendFontSize = 17
+        draw_options.legendFraction = 0.29
+        draw_options.drawMolsSameScale = False
+
+        img = Draw.MolsToGridImage(
+            mols,
+            legends=legends,
+            molsPerRow=2,
+            subImgSize=(250, 250),
+            drawOptions=draw_options,
+        )
+        return img
+
+    @staticmethod
+    def _display_sample_molecules(mols: pd.DataFrame):
+        for _, row in mols.iterrows():
+            with gr.Group():
+                gr.Textbox(value=row["smiles"], label=f"{row['name']} ({row['category']})", interactive=False, scale=3)
+                sample_btn = gr.Button(
+                    f"Load {row['name']}",
+                    scale=1,
+                    size="sm",
+                    variant="primary",
+                )
+                sample_btn.click(
+                    fn=None,
+                    js=f"() => {{window.setJSMESmiles('{row['smiles']}');}}",
+                )
+
+    @staticmethod
+    def clear_all():
+        return "", [], [], None, "Cleared - Draw a new molecule or enter SMILES"
+
+    def handle_search(self, smiles: str, embed_dim: int):
+        if not smiles.strip():
+            return (
+                [],
+                [],
+                None,
+                "Please draw a molecule or enter a SMILES string",
+            )
+        embedding, similar, status = self.molecule_similarity_search_pipeline(smiles, embed_dim)
+        img = self._draw_molecule_grid(similar)
+        return embedding, similar, img, status
+
+    def create_gradio_interface(self):
+        """Create the Gradio interface optimized for JavaScript client usage"""
+        head_scripts = """
+<link rel="preconnect" href="https://jsme-editor.github.io">
+<link rel="preload" href="https://jsme-editor.github.io/dist/jsme/jsme.nocache.js" as="script" crossorigin="anonymous">
+<link rel="preload" href="gradio_api/file=src/main.min.js" as="script">
+<script type="text/javascript" src="https://jsme-editor.github.io/dist/jsme/jsme.nocache.js" crossorigin="anonymous" defer></script>
+<script type="text/javascript" src="gradio_api/file=src/main.min.js" defer></script>
+        """
+
+        with gr.Blocks(
+            title="Chem-MRL: Molecular Similarity Search Demo",
+            theme=gr.themes.Soft(),
+            head=head_scripts,
+        ) as demo:
+            gr.Markdown("""
+            # 🧪 Chem-MRL: Molecular Similarity Search Demo
+
+            Use the JSME editor to draw a molecule or input a SMILES string.
+            The backend encodes the molecule using the Chem-MRL model to produce a vector embedding.
+            Similarity search is performed via an HNSW-indexed Redis vector store to retrieve closest matches.<br/>
+
+            [Model Repo](https://github.com/emapco/chem-mrl) | [Demo Repo](https://github.com/emapco/chem-mrl-demo)
+            """)
+            with gr.Tab("🔬 Molecular Search"), gr.Row():
+                with gr.Column(scale=1):
+                    gr.Markdown("### Molecule Input")
+                    gr.HTML("<div id='jsme_container'></div>")
+
+                    smiles_input = gr.Textbox(
+                        label="SMILES String",
+                        placeholder="Draw a molecule above or enter SMILES here (e.g., CCO for ethanol)",
+                        lines=2,
+                        elem_id="smiles_input",
+                    )
+
+                    embedding_dimension = gr.Dropdown(
+                        choices=SUPPORTED_EMBEDDING_DIMENSIONS,
+                        value=EMBEDDING_DIMENSION,
+                        label="Embedding Dimension",
+                        elem_id="embedding_dimension",
+                    )
+
+                    with gr.Row():
+                        search_btn = gr.Button(
+                            "🔍 Search Molecule Database",
+                            variant="primary",
+                            elem_id="search_btn",
+                        )
+                        clear_btn = gr.Button("🗑️ Clear All", variant="secondary")
+
+                with gr.Column(scale=1):
+                    gr.Markdown("### Search Results")
+                    status_output = gr.Textbox(
+                        label="Status",
+                        interactive=False,
+                        elem_id="status_output",
+                        value="Ready - Draw a molecule or enter SMILES",
+                    )
+
+                    with gr.Accordion("Molecular Embedding Vector", open=False):
+                        embedding_output = gr.JSON(
+                            label="Molecular Embedding",
+                            elem_id="embedding_output",
+                        )
+
+                    with gr.Accordion("Similar Molecules Response", open=False):
+                        similar_molecules_output = gr.JSON(
+                            label="API Response",
+                            elem_id="similar_molecules_output",
+                        )
+
+                    molecule_image = gr.Image(label="Similiar Molecules Grid", type="pil")
+
+            with gr.Tab("📊 Sample Molecules"):
+                gr.Markdown("""
+                ### Sample Molecules in Database
+                Click any button below to load the molecule into the JSME editor:
+                """)
+
+                with gr.Row():
+                    with gr.Column(scale=1):
+                        self._display_sample_molecules(SAMPLE_SMILES[::3])
+                    with gr.Column(scale=1):
+                        self._display_sample_molecules(SAMPLE_SMILES[1::3])
+                    with gr.Column(scale=1):
+                        self._display_sample_molecules(SAMPLE_SMILES[2::3])
+
+            search_btn.click(
+                fn=self.handle_search,
+                inputs=[smiles_input, embedding_dimension],
+                outputs=[
+                    embedding_output,
+                    similar_molecules_output,
+                    molecule_image,
+                    status_output,
+                ],
+                api_name="molecule_similarity_search_pipeline",
+            )
+
+            # Clear UI state
+            clear_btn.click(
+                fn=self.clear_all,
+                js="window.clearJSME",
+                outputs=[
+                    smiles_input,
+                    embedding_output,
+                    similar_molecules_output,
+                    molecule_image,
+                    status_output,
+                ],
+            )
+
+            gr.set_static_paths(paths=["src/"])
+
+        return demo
+
+
+if __name__ == "__main__":
+    app = App()
+    app.demo.launch(**LAUNCH_PARAMETERS)

src/src/constants.py

@@ -0,0 +1,33 @@
+# Model config
+MODEL_NAME = "Derify/ChemMRL-alpha"
+SUPPORTED_EMBEDDING_DIMENSIONS = [1024, 768, 512, 256, 128, 64, 32, 16, 8, 4, 2]
+EMBEDDING_DIMENSION = max(SUPPORTED_EMBEDDING_DIMENSIONS)
+
+# HNSW index parameters
+HNSW_K = 6
+HNSW_PARAMETERS = {
+    # Embedding vector dtype
+    "TYPE": "FLOAT32",
+    # Embedding vectors are normalized so COSINE and IP are equivalent
+    "DISTANCE_METRIC": "IP",
+    # Defines the initial capacity of the vector index. It helps in pre-allocating space for the index.
+    "INITIAL_CAP": 440,
+    # Max number of outgoing edges (connections) for each node in a graph layer.
+    "M": 32,
+    # Max number of connected neighbors to consider during graph building.
+    # Higher values increase accuracy, but also increase index build time.
+    "EF_CONSTRUCTION": 512,
+    # Max top candidates during KNN search. Higher values increase accuracy, but also increase search latency.
+    "EF_RUNTIME": 10,
+}
+
+# Gradio launch parameters
+LAUNCH_PARAMETERS = {
+    "server_name": "0.0.0.0",
+    "server_port": 7860,
+    "share": True,
+    "debug": False,
+    "show_api": False,
+    "pwa": True,
+    "mcp_server": False,
+}

src/src/data.py

@@ -0,0 +1,13 @@
+import os
+
+import pandas as pd
+
+__data_dir = os.path.join(os.path.dirname(__file__), "data")
+
+__dataset_smiles_file = os.path.join(__data_dir, "dataset_smiles.csv")
+__sample_smiles_file = os.path.join(__data_dir, "sample_smiles.csv")
+__isomer_design_subset_file = os.path.join(__data_dir, "isomer_design_subset.csv")
+
+DATASET_SMILES = pd.read_csv(__dataset_smiles_file)
+SAMPLE_SMILES = pd.read_csv(__sample_smiles_file)
+ISOMER_DESIGN_SUBSET = pd.read_csv(__isomer_design_subset_file)

src/src/data/dataset_smiles.csv

@@ -0,0 +1,22 @@
+smiles,name,category
+CCO,Ethanol,Alcohol
+CC(=O)O,Acetic acid,Carboxylic acid
+c1ccccc1,Benzene,Aromatic hydrocarbon
+CC(C)O,Isopropanol,Alcohol
+CCN(CC)CC,Triethylamine,Amine
+c1ccc(cc1)O,Phenol,Aromatic alcohol
+CC(=O)OC1=CC=CC=C1C(=O)O,Aspirin,Carboxylic acid ester
+CN1C=NC2=C1C(=O)N(C(=O)N2C)C,Caffeine,Alkaloid
+CC(C)(C)OC(=O)NC1CCC(CC1)O,Boc-protected cyclohexanol,Protected alcohol
+CCCCCCCCCCCCCCC(=O)O,Palmitic acid,Fatty acid
+c1ccc2c(c1)cccn2,Quinoline,Aromatic heterocycle
+CC1=CC=C(C=C1)C,p-Xylene,Aromatic hydrocarbon
+CCCCO,Butanol,Alcohol
+CC(C)C,Isobutane,Alkane
+c1ccc(cc1)N,Aniline,Aromatic amine
+CC(=O)N,Acetamide,Amide
+CCCCCCCCCCCCCCCCCC(=O)O,Stearic acid,Fatty acid
+c1ccc(cc1)C(=O)O,Benzoic acid,Aromatic carboxylic acid
+CCc1ccccc1,Ethylbenzene,Aromatic hydrocarbon
+CC(C)CC(C)(C)C,"2,2,4-Trimethylpentane",Branched alkane
+O=C1c2ccccc2C(=O)N1C3CCC(=O)NC3=O,Thalidomide,Phthalimides

src/src/data/isomer_design_subset.csv

@@ -0,0 +1,415 @@
+smiles,name,category
+O=C1CCCCC1(c1ccccc1Cl)N1CCCCC1,2-Cl-2'-Oxo-PCP,arylcycloalkylamine
+CCCCNC1(c2ccccc2)CCCCC1,PCB,arylcycloalkylamine
+COc1ccccc1C1(N2CCCCC2)CCCCC1,2-MeO-PCP,arylcycloalkylamine
+c1ccc(C2(n3cccc3)CCCCC2)cc1,PCPyrrole,arylcycloalkylamine
+Fc1cccc(C2(N3CCCC3)CC3CCC2C3)c1,3F-PBCHPy,arylcycloalkylamine
+COc1ccc(C2(NCC3CCCO3)CCCCC2)cc1,1-(4-Methoxyphenyl)-N-[(oxolan-2-yl)methyl]cyclohexan-1-amine,arylcycloalkylamine
+COc1ccc(C2(N3CCCC3)CC3CCC2C3)cc1,4-MeO-PBCHPy,arylcycloalkylamine
+C=CCN(CC)C1(c2ccccc2)CCCCC1,N-Allyl-PCE,arylcycloalkylamine
+CCCCCNC1(c2ccccc2)CCCCC1,PCPe,arylcycloalkylamine
+OC1CCCCC1(c1ccccc1)N1CCCCC1,2'-HO-PCP,arylcycloalkylamine
+Cc1ccccc1C1(N2CCCC(O)C2)CCCCC1,3″-HO-2-Me-PCP,arylcycloalkylamine
+c1ccc(C2(N3CCCCC3)CCCC2)cc1,PCPEP,arylcycloalkylamine
+CCN(C)C1(c2ccccc2C)CCCCC1,2-Me-PCEM,arylcycloalkylamine
+Fc1ccccc1C1(N2CCCC2)CC2CCC1C2,2F-PBCHPy,arylcycloalkylamine
+CC(C)OCCCNC1(c2ccccc2)CCCCC1,1-Phenyl-N-{3-[(propan-2-yl)oxy]propyl}cyclohexan-1-amine,arylcycloalkylamine
+CCNC1(c2ccc(C)cc2)CCCCC1,4-Me-PCE,arylcycloalkylamine
+CCNC1(c2cccc(OC)c2)CCCCC1=O,MXE,arylcycloalkylamine
+CCOC(=O)CCCN(C)C1(c2ccccc2Cl)CCCCC1=O,Et-KE-Bu,arylcycloalkylamine
+CCNC1(c2ccccc2)CCCCC1,PCE,arylcycloalkylamine
+c1cc2c(cc1C1(N3CCOCC3)CCCCC1)OCO2,"3,4-MD-PCMo",arylcycloalkylamine
+OCCNC1(c2ccccc2)CCCCC1,PCEOH,arylcycloalkylamine
+c1ccc(C2(N3CCCCC3)CCCCCCC2)cc1,PCOP,arylcycloalkylamine
+COc1ccc(C2(N3CCCCC3)CC3CCC2C3)cc1,4-MeO-PBCHP,arylcycloalkylamine
+COc1cccc(C2(N3CCCCC3)CCCc3ccccc32)c1,"1-[1-(3-Methoxyphenyl)-1,2,3,4-tetrahydronaphthalen-1-yl]piperidine",arylcycloalkylamine
+CCNC12CCCCC1CCc1ccccc12,"N-Ethyl-1,3,4,9,10,10a-hexahydrophenanthren-4a(2H)-amine",arylcycloalkylamine
+c1ccc([C@]2(N3CCCCC3)C[C@H]3C[C@H]3C2)cc1,"1-[(1R,3r,5S)-3-Phenylbicyclo[3.1.0]hexan-3-yl]piperidine",arylcycloalkylamine
+c1ccc(C2(N3CCCCC3)CN3CCC2CC3)cc1,P3QP,arylcycloalkylamine
+COc1cccc(C2(N3CCCC3)CCCCC2=O)c1,3-MeO-2'-Oxo-PCPy,arylcycloalkylamine
+c1cc2c(cc1C1(N3CCCCC3)CCCCC1)OCCO2,"3,4-ED-PCP",arylcycloalkylamine
+NC1(c2ccccc2Cl)CCCNC1=O,3-Amino-3-(2-chlorophenyl)piperidin-2-one,arylcycloalkylamine
+CC[C@]1(c2ccccc2)OC[C@H]([C@@H]2CCCCN2)O1,CL-1848C,arylcycloalkylamine
+O=[N+]([O-])c1cccc(C2(N3CCCCC3)CCCCC2)c1,3-NO2-PCP,arylcycloalkylamine
+Fc1cccc(C2(N3CCCCC3)CCCCCC2)c1,3F-PCHEP,arylcycloalkylamine
+CCNC1(c2cccc(OC(C)C)c2)CCCCC1=O,iPXE,arylcycloalkylamine
+COc1cc(C2(N3CCCCC3)CCCCC2)ccc1C,3-MeO-4-Me-PCP,arylcycloalkylamine
+O=C1CCC(c2ccccc2)(N2CCCCC2)CC1,4'-Oxo-PCP,arylcycloalkylamine
+CCNC1(c2ccc3c(c2)OCO3)CCCCC1,"3,4-MD-PCE",arylcycloalkylamine
+CNC1(c2ccccc2OC)CCCCC1=O,2-MDCK,arylcycloalkylamine
+CCOc1cccc(C2(N3CCCCC3)CCCCC2)c1,3-EtO-PCP,arylcycloalkylamine
+CNC1(c2ccccc2C)CCCCC1=O,o-MDCK,arylcycloalkylamine
+CNC1(c2cccc(Cl)c2)CCCCC1=O,meta-Ketamine,arylcycloalkylamine
+CCNC1(c2cc(OC)ccc2Br)CCCCC1=O,Bromo-methoxetamine,arylcycloalkylamine
+CNC1(c2cccc(OC)c2)CCCCC1=O,MXM,arylcycloalkylamine
+CCCOC(=O)CCNC1(c2ccccc2Cl)CCCCC1=O,nPr-NKE-Et,arylcycloalkylamine
+CC1CCCCC1(c1ccccc1)N1CCCCC1,2'-Me-PCP,arylcycloalkylamine
+Cc1ccc(C2(N3CCC(O)CC3)CCCCC2O)cc1,"2',4″-HO-4-Me-PCP",arylcycloalkylamine
+NC1(c2ccccc2Cl)CCC(O)CC1=O,"(2,5)-HNK",arylcycloalkylamine
+Fc1cccc(C2(N3CCCCC3)CC3CCC2C3)c1,3F-PBCHP,arylcycloalkylamine
+CC(C)NC1(c2ccccc2)CCCCC1,PCiP,arylcycloalkylamine
+Fc1ccc(C2(N3CCCCC3)CCCCCC2)cc1,4F-PCHEP,arylcycloalkylamine
+CCCNC1(c2ccccc2)CCCC1,PCPEPr,arylcycloalkylamine
+CC(C)Oc1cccc(C2(N)CCCCC2)c1,3-IpO-PCA,arylcycloalkylamine
+CC(O)CNC1(c2ccccc2)CCCCC1,1-[(1-Phenylcyclohexyl)amino]propan-2-ol,arylcycloalkylamine
+Cc1ccc(C2(N(C)C)CCC(O)(CCc3ccccc3)CC2)cc1,MDPC,arylcycloalkylamine
+Cc1cc(O)ccc1C1(N2CCOCC2)CCCCC1,4-HO-2-Me-PCMo,arylcycloalkylamine
+O=C1CCCCC1(c1ccccc1)N1CCCC1,2-Phenyl-2-(pyrrolidin-1-yl)cyclohexan-1-one,arylcycloalkylamine
+CNC1(c2ccccc2Cl)CCCCCC1=O,2-(2-Chlorophenyl)-2-(methylamino)cycloheptan-1-one,arylcycloalkylamine
+CCN(CC)C1(c2ccccc2)CCCCC1,PCDE,arylcycloalkylamine
+COc1cccc([C@@]2(O)CCN(C3(c4ccccc4)CCCCC3)C[C@@H]2CN(C)C)c1,US 7049327 #6,arylcycloalkylamine
+Cc1cccc(C2(N3CCCC3)CC3CCC2C3)c1,3-Me-PBCHPy,arylcycloalkylamine
+C#Cc1ccc2c(c1)C(c1ccccc1F)=N[C@@H](C)c1c(C(=O)N(C)C)ncn1-2,GL-i-54,aryldiazepine
+O=C1CN=C(c2c(F)cccc2F)c2cc(Cl)ccc2N1,Ro-07-3953,aryldiazepine
+O=C1CN=C(c2ccn[nH]2)c2cc(Cl)ccc2N1,SCHEMBL9684958,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccn1)=NCc1c(C(N)=O)ncn1-2,MP-ii-065,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2c(Cl)cccc21,Ro-13-0699,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccc1)=NCc1c(C(=O)OCC(F)(F)F)ncn1-2,DM-ii-20,aryldiazepine
+O=C1OCCC2N=C(c3ccccc3)c3ccccc3-n3cnc1c32,"8-Phenyl-6,6a-dihydro-4-oxa-2,7,12b-triazabenzo[b]cyclopenta[kl]heptalen-3(5H)-one",aryldiazepine
+CCC1C(C)=NN=C(c2ccc(OC)c(OC)c2)c2cc(OC)c(OC)cc21,Tofisopam,aryldiazepine
+CN1C(=O)CN=C(c2ccc(F)cc2)c2cc(Cl)ccc21,4F-Diazepam,aryldiazepine
+CC(C)COC(=O)c1ncn2c1C1CCCN1C(=O)c1c(Cl)cccc1-2,Ro-16-7082,aryldiazepine
+CN1C(=O)CN=C(c2cccs2)c2cc(Cl)ccc21,WZ-113,aryldiazepine
+CC(=O)c1ccc2c(c1)C(c1ccccc1)=NCC(=O)N2,Ro-20-3053,aryldiazepine
+C#Cc1ccc2c(c1)C(=O)N(C)Cc1c(C(=O)OC(C)(C)C)ncn1-2,RY-24,aryldiazepine
+CC(C)(C)OC(=O)c1ncn2c1C1CCCN1C(=O)c1c(F)cccc1-2,Ro-16-8912,aryldiazepine
+O=C1CN=C(c2ccccc2F)c2cc(Cl)ccc2N1CCO,Ro-07-2750,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2Cl)c2cc([N+](=O)[O-])ccc21,Ro-05-4082,aryldiazepine
+Cc1nnc2n1-c1sc(Cl)cc1C(c1ccccc1)=NC2,Deschloroclotizolam,aryldiazepine
+CCOC(=O)c1ncn2c1C(C)N=C(c1ccc(F)cc1)c1cc(F)ccc1-2,"Ethyl 8-fluoro-6-(4-fluorophenyl)-4-methyl-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate",aryldiazepine
+O=C1CN=C(c2ccccc2F)c2cc(Cl)c(Cl)cc2N1,Ro-20-8065,aryldiazepine
+CN1C(=O)CN=C(c2c(F)cccc2F)c2cc(Br)ccc21,Ro-13-3780,aryldiazepine
+CN1C(=O)CC(=O)N(c2ccccc2)c2cc(Cl)ccc21,Clobazam,aryldiazepine
+OCc1nnc2n1-c1ccc(Cl)cc1C(c1ccccc1)=NC2,alpha-Hydroxyalprazolam,aryldiazepine
+O=C1Nc2ccc(Cl)cc2C(c2ccccc2)=NC1O,Oxazepam,aryldiazepine
+CCOC(=O)c1ncn2c1CN=C(c1ccc([N+](=O)[O-])cc1)c1ccccc1-2,Ro-23-2896,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2cc(Br)ccc21,Flubrometazepam,aryldiazepine
+CN(C)C(=O)c1nc2n(n1)-c1ccc(Cl)cc1C(c1ccccc1Cl)=NC2,Rilmazolam,aryldiazepine
+CC1N=C(c2ccccc2F)c2cc(Cl)ccc2NC1=O,Ro-11-4878,aryldiazepine
+CCOC(=O)c1ncn2c1C1CCCN1C(=O)c1cc(Br)ccc1-2,RY-054,aryldiazepine
+CCOC(=O)c1ncn2c1C1CCN1C(=O)c1c(Cl)cccc1-2,Ro-16-0858,aryldiazepine
+CN1C(=O)CN=C(c2ccc(Cl)cc2)c2cc(Cl)ccc21,Ro-05-4864,aryldiazepine
+CN(C)CCc1nnc2n1-c1ccc(Cl)cc1C(c1ccccc1)=NC2,U-43465F,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2c(Cl)cc(Cl)cc21,Ro-13-0882,aryldiazepine
+CCc1ccc2c(c1)C(c1ccccc1)=NCC(=O)N2,Ro-20-2533,aryldiazepine
+CCOC(=O)c1ncn2c1CC=C(c1ccccc1F)c1ccccc1-2,Ro-15-2200,aryldiazepine
+O=C1CC=C(c2ccccc2Cl)c2cc([N+](=O)[O-])ccc2N1,Ro-15-8867,aryldiazepine
+Cc1cc2c(cc1Cl)NC(=O)CN=C2c1ccccc1F,Ro-20-8552,aryldiazepine
+N#Cc1ncn2c1CN=C(c1ccc(Cl)cc1)c1cc(Cl)ccc1-2,"8-Chloro-6-(4-chlorophenyl)-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carbonitrile",aryldiazepine
+CC(C)c1nc(-c2ncn3c2CN(C)C(=O)c2c(Cl)cccc2-3)no1,FG-8205,aryldiazepine
+Cc1nnc2n1-c1ccc([N+](=O)[O-])cc1C(c1ccccc1Cl)=NC2,Clonazolam,aryldiazepine
+O=C1CN=C(c2ccccc2[N+](=O)[O-])c2cc(Br)ccc2N1,QH-II-067a,aryldiazepine
+NC(=O)c1ncn2c1CC=C(c1ccccc1F)c1cc(Cl)ccc1-2,Ro-14-3930,aryldiazepine
+COC(=O)CC[C@@H]1N=C(c2ccccn2)c2cc(Br)ccc2-n2c(C)cnc21,CNS-7056,aryldiazepine
+CN1C(=O)CN=C(C2CCC2)c2cc(F)ccc21,"5-Cyclobutyl-7-fluoro-1-methyl-1,3-dihydro-2H-1,4-benzodiazepin-2-one",aryldiazepine
+O=C1CN=C(c2c(F)cccc2Cl)c2cc(Cl)ccc2N1,Ro-07-5193,aryldiazepine
+CN1C(=O)CN=C(c2ccccc2F)c2cc(I)ccc21,Ro-07-9957,aryldiazepine
+O=C1CN=C(c2ccccc2)c2cc(Cl)ccc2N1CC1CC1,Prazepam,aryldiazepine
+CCOC(=O)c1ncn2c1CN=C(c1cccc([N+](=O)[O-])c1)c1cc(F)ccc1-2,"Ethyl 8-fluoro-6-(3-nitrophenyl)-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate",aryldiazepine
+Clc1ccc2c(c1)-c1ncnn1Cc1c(Cl)ncn1-2,Ro4882224,aryldiazepine
+[N-]=[N+]=Nc1ccc2c(c1)C(c1ccccc1F)=NCC(=O)N2,Ro-14-3074,aryldiazepine
+Fc1ccccc1C1=NCC=Cc2ccc(Cl)cc21,Ro-22-1274,aryldiazepine
+Nc1ncc2c(n1)-c1ccc(Cl)cc1C(c1ccccc1F)=NC2,Ro-22-2038,aryldiazepine
+Clc1ccc2c(c1)C(c1ccccc1)=NCc1nncn1-2,Estazolam,aryldiazepine
+Cc1nn(C)c2c1C(c1ccccc1F)=NCC(=O)N2C,Zolazepam,aryldiazepine
+Cc1nnc2n1-c1ccc(Cl)cc1C(c1ccccn1)=NC2,Pyclazolam,aryldiazepine
+S=C1CN=C(c2ccccc2)c2cc(Cl)ccc2N1,Thionordazepam,aryldiazepine
+O=C1CN=C(c2ccccc2)c2cc(Cl)ccc2N1CC(F)(F)F,Halazepam,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccc1)=NCC1=N/C(=C\N3CCN(C)CC3)C(=O)N12,PS-i-37,aryldiazepine
+C#Cc1ccc2c(c1)C(c1ccccc1F)=NCc1c(C(=O)OCC(F)(F)F)ncn1-2,JY-i-59,aryldiazepine
+O=C1CN=C(c2ccccc2F)c2cc([N+](=O)[O-])ccc2N1,Fonazepam,aryldiazepine
+CN1CCN(/C=C2\N=C3CN=C(c4ccccc4)c4cc(Br)ccc4N3C2=O)CC1,PS-1-35,aryldiazepine
+CCCCCCC(C)(C)c1ccc([C@@H]2CCC[C@H](O)C2)c(O)c1,"(1R,3S) CP 47,497",cannabinoid
+CCCc1ccc(C(=O)c2cn(CCCCCF)c3ccccc23)c2ccccc12,SGT-62,cannabinoid
+CC(C)[C@H](NC(=O)c1cn(Cc2ccc(F)cc2)c2ccccc12)C(N)=O,AB-FUBICA,cannabinoid
+CCCCCn1nc(C(=O)N[C@@H](Cc2ccccc2)C(=O)OC)c2ccccc21,MPP-PINACA,cannabinoid
+O=C(Cc1cn(Cc2ccc(F)cc2)c2ccccc12)NC1C2CC3CC(C2)CC1C3,AFUBIATA,cannabinoid
+CC(C)(C)c1nc(N2CCC(F)(F)C2)c2nnn(Cc3ccccc3S(C)(=O)=O)c2n1,"5-tert-Butyl-7-(3,3-difluoropyrrolidin-1-yl)-3-{[2-(methanesulfonyl)phenyl]methyl}-3H-[1,2,3]triazolo[4,5-d]pyrimidine",cannabinoid
+CCC(C(N)=O)N(C)C(=O)c1nn(Cc2ccc(F)cc2)c2ccccc12,AB-FUBINACA isomer 5,cannabinoid
+COc1cccc2c3c(=O)n([C@@H]4C(C)(C)[C@@H]5CC[C@@]4(C)C5)ccc3n(Cc3ccncc3)c12,WO 2001/058869 #190,cannabinoid
+CCCC(F)Cn1cc(C(=O)C2C(C)(C)C2(C)C)c2ccccc21,(2-Fluoropentyl) XLR-11,cannabinoid
+O=C(Nc1ccccc1)c1cn(CCCCCF)c2ccccc12,LTI-701,cannabinoid
+C=CCCCn1nc(C(=O)N[C@H](C(=O)OC)C(C)C)c2ccccc21,MMB-4en-PINACA,cannabinoid
+O=C(Nc1ccccc1)c1nn(CC2CCOCC2)c2ccccc12,SGT-224,cannabinoid
+Cc1ccc(C(=O)NC(C)(C)c2ccccc2)cc1S(=O)(=O)N1CCC(F)(F)CC1,SGT-233,cannabinoid
+CCCCCn1nc(C(=O)c2ccc(OC)c3ccccc23)c2ccccc21,SGT-35,cannabinoid
+CC1(C)C(C(=O)c2cn(CC3CCCCC3)c3ccccc23)C1(C)C,(CHM) UR-144,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2ccc(N=C=S)c3ccccc23)c2ccc(N=C=S)cc21,AM-2215,cannabinoid
+CCCCCCC1(c2ccc(-c3cc(C)cc(C)c3)c(O)c2)CC1,"4-(1-Hexylcyclopropyl)-3',5'-dimethyl[1,1'-biphenyl]-2-ol",cannabinoid
+COC(=O)[C@@H](NC(=O)c1cn(Cc2ccc(F)cc2)c2ncccc12)C(C)(C)C,MDMB-FUB7AICA,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2ccccc2I)c2ccccc21,AM-2233,cannabinoid
+O=C(c1cccc2ccccc12)n1c(-c2ccncc2)nc2ccccc21,(Naphthalen-1-yl)[2-(pyridin-4-yl)-1H-benzimidazol-1-yl]methanone,cannabinoid
+C=CCCCn1nc(C(=O)N[C@H](C(=O)OC)C(C)(C)C)c2ccccc21,MDMB-4en-PINACA,cannabinoid
+Oc1ccc(/C=C2\C=C(CCN3CCOCC3)c3ccccc32)c2ccccc12,US 5292736 #19,cannabinoid
+CCc1c(C(=O)NN2CCCCC2)nn(-c2ccc(Cl)cc2Cl)c1-c1ccc(C#Cc2ccc(C(F)(F)F)cc2)s1,TM-38837,cannabinoid
+COC(=O)[C@H](Cc1ccccc1)NC(=O)c1cn(CCCCCF)c2ccccc12,MPHP-2201,cannabinoid
+Cc1ccc2c(C(C[N+](=O)[O-])c3ccccc3)c(-c3ccccc3)[nH]c2c1,GAT211,cannabinoid
+CCCCCn1nc(C(=O)NC2CCCCC2)c2ccccc21,SGT-85,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2ccc([N+](=O)[O-])c3ccccc23)c2ccccc21,AM-1299,cannabinoid
+CCc1c(C)cc(C(=O)NC2(C(=O)O)CCCCC2)c(=O)n1Cc1ccc(F)cc1,S-777469,cannabinoid
+CCCCCC(C)NC(=O)c1nn(CCCCC)c2ccccc12,SGT-71,cannabinoid
+CC1(C)Oc2cc(C34CC5CC(CC(C5)C3)C4)cc(O)c2[C@@H]2CC(O)CC[C@H]21,"(6aR,10aR)-3-(Adamantan-1-yl)-6,6-dimethyl-6a,7,8,9,10,10a-hexahydro-6H-dibenzo[b,d]pyran-1,9-diol",cannabinoid
+C=C(C)[C@H]1CCC(CO)=C[C@@H]1c1c(O)cc(C(C)(C)CCCCN2CCOCC2)cc1O,"(1'S,2'S)-5'-(Hydroxymethyl)-4-[2-methyl-6-(morpholin-4-yl)hexan-2-yl]-2'-(prop-1-en-2-yl)-1',2',3',4'-tetrahydro[1,1'-biphenyl]-2,6-diol",cannabinoid
+O=C(N/N=C1\C(=O)N(CCCCCF)c2ccccc21)c1ccccc1,BZO-5F-POXIZID,cannabinoid
+CCCCCOc1c(OC)ccc2cc(C(=O)NCc3ccc4c(c3)OCO4)c(=O)[nH]c12,JTE-907,cannabinoid
+CCCCn1nc(C(=O)N[C@@H](Cc2ccccc2)C(=O)OC)c2ccccc21,MPP(N)-073,cannabinoid
+CCCCCCC(C)(C)c1ccc([C@@H]2CCC[C@@H](O)C2)c(O)c1,"(R,R)-epi-CP 47,497",cannabinoid
+CCCCCOc1ccc(C(=O)c2cccc3ccccc23)c2ccccc12,CB-13,cannabinoid
+COc1ccccc1OC(=O)c1ccc(C)c(S(=O)(=O)N2CCOCC2)c1,2-Methoxyphenyl 4-methyl-3-(morpholine-4-sulfonyl)benzoate,cannabinoid
+CCCCCC#Cc1cc(O)c2c(c1)OC(C)(C)[C@@H]1CC=C(C)C[C@@H]21,AMG-1,cannabinoid
+CC(C)[C@H](NC(=O)c1cn(CCCCCF)c2ccccc12)C(N)=O,5F-ABICA,cannabinoid
+CCCCCn1cc(C(=O)NC23CC4CC(CC(C4)C2)C3)c(=O)c2ccc(F)cc21,"N-(Adamantan-1-yl)-7-fluoro-4-oxo-1-pentyl-1,4-dihydroquinoline-3-carboxamide",cannabinoid
+CC1=CC[C@@H]2[C@@H](C1)c1c(O)cc(C(C)(C)CCCCCC#N)cc1OC2(C)C,O-774,cannabinoid
+CC(C)(C)[C@H](NC(=O)n1c(=O)n(CCN2CCOCC2)c2ccccc21)C(N)=O,WO 2008/032164 #4,cannabinoid
+COC(=O)[C@H](Cc1ccccc1)NC(=O)c1cn(CC2CCCCC2)c2ccccc12,MPP-CHMICA,cannabinoid
+Fc1ccc2nc(-c3ccccn3)n(Cc3cccc4ccccc34)c2c1,6-Fluoro-1-[(naphthalen-1-yl)methyl]-2-(pyridin-2-yl)-1H-benzimidazole,cannabinoid
+CCCCCn1cc(C(=O)N[C@@H](C)c2ccccc2)c2ccccc21,1-Pentyl-N-[(1S)-1-phenylethyl]-1H-indole-3-carboxamide,cannabinoid
+CCCCCCCC(C)c1cc(OC(=O)CCCN(CC)CC)c2c(c1)OC(C)(C)C1=C2CC(C)CC1,SP-325,cannabinoid
+CCCCCn1cc(C(=O)c2cccc3ccccc23)cc1-c1cccc(C)c1,JWH-346,cannabinoid
+O=C(Cc1ccccc1)N/N=C1\C(=O)N(CC2CCCCC2)c2cc(OCCN3CCOCC3)ccc21,"N'-{1-(Cyclohexylmethyl)-6-[2-(morpholin-4-yl)ethoxy]-2-oxo-1,2-dihydro-3H-indol-3-ylidene}-2-phenylacetohydrazide",cannabinoid
+Cc1c(C(=O)C2C(C)(C)C2(C)C)c2ccccc2n1CCCCCF,2-Me-XLR-11,cannabinoid
+COc1ccc2c(c1)CCC2CNC(=O)c1cn(CCCCCF)c2ccccc12,SGT-105,cannabinoid
+CCCCCc1cc(O)c2c(c1)OC(C)(C)C(CCC(C)=O)C2=O,CBCN,cannabinoid
+CCCCCn1nc(C(=O)NC(C)(C)c2ccc(F)cc2)c2ccccc21,SGT-60,cannabinoid
+CC(C)(C)[C@H](NC(=O)c1nn(CC2CCCCCC2)c2ccccc12)C(=O)NCCO,WO 2009/106980 #433,cannabinoid
+CC(C)(NC(=O)c1cn(Cc2ccc(F)cc2)c2ccccc12)c1ccccc1,CUMYL-FUBICA,cannabinoid
+COC(=O)[C@@H](NC(=O)c1nn(CC2CCCCC2)c2ccccc12)C(C)(C)C,MDMB(N)-CHM,cannabinoid
+CCCCCn1nc(C(=O)c2cccc3ccccc23)c2ccccc21,SGT-17,cannabinoid
+CCCCCn1cc(C(=O)N[C@H](C(N)=O)C(C)(C)C)c2ccccc21,ADBICA,cannabinoid
+CN1CCCCC1Cn1cc(C(=O)c2cc(N=C=S)ccc2I)c2ccc(N=C=S)cc21,AM-2223,cannabinoid
+CCCCCCCCCn1cc(C(=O)c2cccc3ccccc23)c2ccccc21,(Nonyl) JWH-018,cannabinoid
+CCCCCCC(C)(C)c1cc(OC)c([C@@H]2C=C(CO)C3CC2C3(C)C)c(OC)c1,HU-433,cannabinoid
+C=CCc1cc(OC)c(OC)c(OC)c1OC,"2,3,4,5-Tetramethoxyphenylprop-2-ene",essential oil
+C/C=C\c1ccc(O)c(OC)c1,cis-4-Hydroxy-3-methoxyphenylprop-1-ene,essential oil
+C/C=C/c1ccc2c(c1OC)OCO2,Isocroweacin,essential oil
+C/C=C\c1cc(OC)c(O)c(OC)c1,"cis-4-Hydroxy-3,5-dimethoxyphenylprop-1-ene",essential oil
+C=CCc1ccc2c(c1OC)OCO2,Croweacin,essential oil
+C/C=C\c1cc2c(cc1OC)OCO2,"cis-2-Methoxy-4,5-methylenedioxyphenylprop-1-ene",essential oil
+C=CCc1ccc2c(c1)OCO2,Safrole,essential oil
+C/C=C\c1cc(OC)c2c(c1)OCO2,cis-Isomyristicin,essential oil
+C/C=C\c1ccc(OC)cc1OC,"cis-2,4-Dimethoxyphenylprop-1-ene",essential oil
+C/C=C\c1ccc2c(c1OC)OCO2,"cis-2-Methoxy-3,4-methylenedioxyphenylprop-1-ene",essential oil
+C/C=C/c1ccc(OC)c(OC)c1,Methylisoeugenol,essential oil
+C=CCc1ccc(OC)cc1OC,Osmorrhizole,essential oil
+C=CCc1ccc(OC)c(OC)c1,Methyleugenol,essential oil
+C=CCc1ccc(O)c(OC)c1,Eugenol,essential oil
+C=CCc1cc2c(c(OC)c1OC)OCO2,Dillapiole,essential oil
+C/C=C\c1ccc2c(c1)OCO2,alpha-Isosafrole,essential oil
+C/C=C/c1cc(OC)c(OC)c(OC)c1,Isoelemicin,essential oil
+C=CCc1ccc(OC)c(O)c1,Chavibetol,essential oil
+C=CCc1cc(OC)c(OC)c(OC)c1,Elemicin,essential oil
+C=CCc1cc(OC)cc(OC)c1,"3,5-Dimethoxyphenylprop-2-ene",essential oil
+C/C=C\c1cc(OC)c(OC)c(OC)c1,cis-Isoelemicin,essential oil
+C/C=C/c1cc(OC)c(OC)c(OC)c1OC,"trans-2,3,4,5-Tetramethoxyphenylprop-1-ene",essential oil
+C/C=C/c1cc(OC)c2c(c1OC)OCO2,Isoapiole,essential oil
+C/C=C\c1ccc(OC)c(O)c1,cis-3-Hydroxy-4-methoxyphenylprop-1-ene,essential oil
+C=CCc1ccc(OC)cc1,Estragole,essential oil
+C/C=C\c1cc(OC)c2c(c1OC)OCO2,"cis-2,5-Dimethoxy-3,4-methylenedioxyphenylprop-1-ene",essential oil
+C=CCc1cc(OC)c2c(c1OC)OCO2,Apiole,essential oil
+C/C=C/c1ccc(OC)cc1OC,Nothosmyrnol,essential oil
+C/C=C/c1cc(OC)cc(OC)c1,"trans-3,5-Dimethoxyphenylprop-1-ene",essential oil
+C/C=C\c1cc2c(c(OC)c1OC)OCO2,"cis-2,3-Dimethoxy-4,5-methylenedioxyphenylprop-1-ene",essential oil
+C/C=C\c1ccc(OC)cc1,cis-4-Methoxyphenylprop-1-ene,essential oil
+C/C=C/c1cc(OC)c(O)c(OC)c1,"2,6-Dimethoxy-4-[(1E)-prop-1-en-1-yl]phenol",essential oil
+C=CCc1cc(OC)c(OC)cc1OC,γ-Asarone,essential oil
+C/C=C/c1ccc(OC)cc1,Anethole,essential oil
+C=CCc1cc2c(cc1OC)OCO2,Asaricin,essential oil
+C/C=C/c1cc2c(c(OC)c1OC)OCO2,Isodillapiole,essential oil
+C/C=C/c1cc(OC)c(OC)cc1OC,Asarone,essential oil
+C/C=C/c1ccc(O)c(OC)c1,Isoeugenol,essential oil
+C/C=C\c1cc(OC)cc(OC)c1,"cis-3,5-Dimethoxyphenylprop-1-ene",essential oil
+C/C=C\c1cc(OC)c(OC)cc1OC,beta-Asarone,essential oil
+C=CCc1cc(OC)c2c(c1)OCO2,Myristicin,essential oil
+C/C=C/c1ccc(OC)c(O)c1,2-Methoxy-5-[(1E)-prop-1-en-1-yl]phenol,essential oil
+C/C=C/c1cc(OC)c2c(c1)OCO2,Isomyristicin,essential oil
+C/C=C\c1ccc(OC)c(OC)c1,"cis-3,4-Dimethoxyphenylprop-1-ene",essential oil
+C=CCc1cc(OC)c(O)c(OC)c1,5-Methoxyeugenol,essential oil
+C/C=C/c1ccc2c(c1)OCO2,beta-Isosafrole,essential oil
+C/C=C\c1cc(OC)c(OC)c(OC)c1OC,"cis-2,3,4,5-Tetramethoxyphenylprop-1-ene",essential oil
+C/C=C/c1cc2c(cc1OC)OCO2,Carpacin,essential oil
+CCC(=O)N(c1ccccc1)C1(COC)CCN(CC(O)c2cccs2)CC1,beta-Hydroxy-sufentanil,fentanyl
+CCC(=O)N(CCc1ccccc1)C1CCN(CCc2ccccc2)CC1,(ethylene) Fentanyl,fentanyl
+CCC(=O)N(c1ccccc1F)C1(c2ccccc2)CCN(CCn2cccn2)CC1,N-(2-Fluorophenyl)-N-{4-phenyl-1-[2-(1H-pyrazol-1-yl)ethyl]piperidin-4-yl}propanamide,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(C(C)Cc2cc(OC)c(I)cc2OC)CC1,N-(DOI) fentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1(COC(=O)C(C)(C)C)CCN(CCN2C(=O)c3ccccc3C2=O)CC1,"{1-[2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)ethyl]-4-[phenyl(propanoyl)amino]piperidin-4-yl}methyl 2,2-dimethylpropanoate",fentanyl
+CC(C)C(=O)N(c1cccc(F)c1)C1CCN(CCc2ccccc2)CC1,m-Fluoro-isobutyrylfentanyl,fentanyl
+CCCC(=O)N(c1ccc(OC)cc1)C1CCN(CCc2ccccc2)CC1,p-MeO-BF,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(Cc2ccccc2)CC1,Benzylcarfentanil,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCc2ccsc2)CC1,Methyl 4-[phenyl(propanoyl)amino]-1-[2-(thiophen-3-yl)ethyl]piperidine-4-carboxylate,fentanyl
+COCC(=O)N(c1ccc(C)cc1)C1CCN(CCc2ccccc2)CC1,p-Methyl-methoxyacetylfentanyl,fentanyl
+CCCCC(=O)N(c1ccc(OC)cc1)C1CCN(CCc2ccccc2)CC1,p-Methoxy-valerylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)CC1F,NFEPP,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(C(C)Cc2cc(OC)c(Br)cc2OC)CC1,N-(DOB) fentanyl,fentanyl
+CCCC(=O)N(c1ccccc1)C1CCN(C)CC1,N-Methyl-butyrylfentanyl,fentanyl
+O=C(c1ccco1)N(c1ccccc1)C1CCN(CCc2ccc(O)cc2)CC1,4'-Hydroxy-furanylfentanyl,fentanyl
+CCC(=O)OCC1(N(C(=O)CC)c2ccccc2)CCN(CCN2C(=O)c3ccccc3C2=O)CC1,"{1-[2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)ethyl]-4-[phenyl(propanoyl)amino]piperidin-4-yl}methyl propanoate",fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)C(C)C1,2-Methylfentanyl,fentanyl
+CC(=O)N(c1ccccc1)C1CCN(CCCc2ccccc2)CC1,Phenylpropyl-acetyl-norfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCOc2ccccc2)CC1,POE-F,fentanyl
+O=C(c1ccco1)N(c1ccccc1)C1CCNCC1,Furanyl-norfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(Cc2ccccc2)CC1,R-4129,fentanyl
+Cc1ccc(NC2CCN(CCc3ccccc3)CC2)cc1,p-Me-4-ANPP,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)O)CCN(CCc2ccccc2)C(O)C1,2-Hydroxycarfentanil acid,fentanyl
+COCC(=O)N(c1ccc(F)cc1)C1CCN(CCc2ccccc2)CC1,p-Fluoro-methoxyacetylfentanyl,fentanyl
+Cc1cccc(N(C(=O)c2ccco2)C2CCN(CCc3ccccc3)CC2)c1,m-Methyl-furanylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2F)CC1,2″-Fluorofentanyl,fentanyl
+COc1ccccc1N(C(=O)c1ccco1)C1CCN(CCc2ccccc2)CC1,o-Methoxy-furanylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CC(O)c2ccc(F)cc2)CC1,N-{1-[2-(4-Fluorophenyl)-2-hydroxyethyl]piperidin-4-yl}-N-phenylpropanamide,fentanyl
+CC(=O)N(c1ccccc1)C1CCN(CCc2cccs2)CC1,Acetylthiofentanyl,fentanyl
+COCC(=O)N(c1cccc(F)c1)C1CCN(CCc2ccccc2)CC1,m-Fluoro-methoxyacetylfentanyl,fentanyl
+COc1cc(CCN2CCC(N(C(=O)C(=O)CCC(=O)O)c3ccccc3)CC2)ccc1O,N-Phenyl-N-[1-(2-phenylethyl)piperidin-4-yl]-2-oxo-4-carboxybutanamide,fentanyl
+CCC(=O)N(c1ccccc1)C1(COC)CCN(CCn2nnn(CC)c2=O)CC1,R-39209,fentanyl
+CCC(=O)N(c1ccc(Cl)cc1)C1CCN(CCc2ccccc2)CC1,p-CF,fentanyl
+CCC(=O)N(c1ccccc1)C1CCNCC1,Norfentanyl,fentanyl
+COC(=O)CCN1CCC(Nc2ccccc2)(C(=O)OC)CC1,Despropionyl remifentanil,fentanyl
+CCC(=O)N(c1ccccc1)C1(c2nc(C)cs2)CCN(CCc2ccccc2)CC1,"N-[4-(4-Methyl-1,3-thiazol-2-yl)-1-(2-phenylethyl)piperidin-4-yl]-N-phenylpropanamide",fentanyl
+CCC(=O)N(c1ccccc1)C1(c2ccccc2)CCN(CCc2ccccc2)CC1,4-Phenylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1C)C1CCN(CCc2ccccc2)CC1,o-Methylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCCCCF)CC1,Fluoropentyl-norcarfentanil,fentanyl
+O=C(c1ccco1)N(c1ccccc1)C1CCN(Cc2ccccc2)CC1,Furanylbenzylfentanyl,fentanyl
+CCC(=O)N(c1ccc2ncoc2c1)C1CCN(CCc2ccccc2)CC1,N-Benzoxazolyl-fentanyl,fentanyl
+COCC(=O)N(c1cccc(C)c1)C1CCN(CCc2ccccc2)CC1,m-Methyl-methoxyacetylfentanyl,fentanyl
+CC(C)C(=O)OCC1(N(C(=O)C(C)C)c2ccccc2)CCN(CCN2C(=O)c3ccccc3C2=O)CC1,"{1-[2-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)ethyl]-4-[(2-methylpropanoyl)(phenyl)amino]piperidin-4-yl}methyl 2-methylpropanoate",fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCc2ccccn2)CC1,Methyl 4-[phenyl(propanoyl)amino]-1-[2-(pyridin-2-yl)ethyl]piperidine-4-carboxylate,fentanyl
+c1ccc(CCN2CCC(Nc3ccccc3)CC2)cc1,ANPP,fentanyl
+CCCc1cc(OC)c(CCN2CCC(N(C(=O)CC)c3ccccc3)CC2)cc1OC,N-(2C-P) fentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CC(O)c2ccncc2)CC1C,N-{1-[2-Hydroxy-2-(pyridin-4-yl)ethyl]-3-methylpiperidin-4-yl}-N-phenylpropanamide,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CC(=O)c2cccc(F)c2)CC1,N-{1-[2-(3-Fluorophenyl)-2-oxoethyl]piperidin-4-yl}-N-phenylpropanamide,fentanyl
+CCCCc1cc(OC)c(CC(C)N2CCC(N(C(=O)CC)c3ccccc3)CC2)cc1OC,N-(DOBU) fentanyl,fentanyl
+COCC(=O)N(c1ccccc1)C1C(C)CN(CCn2cccn2)CC1C,"N-{3,5-Dimethyl-1-[2-(1H-pyrazol-1-yl)ethyl]piperidin-4-yl}-2-methoxy-N-phenylacetamide",fentanyl
+Cc1ccc(N(C(=O)C2CCCO2)C2CCN(CCc3ccccc3)CC2)cc1,p-Methyl-tetrahydrofuranylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2cc(OC)cc(OC)c2)CC1,"3″,5″-DiMeO-F",fentanyl
+CC(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)CC1,Acetylfentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(CCc2ccc(F)cc2)CC1,4″-Fluorofentanyl,fentanyl
+CCC(=O)N(c1ccccc1)C1(COC(=O)OC)CCN(CCc2cccs2)CC1,Methyl {4-[phenyl(propanoyl)amino]-1-[2-(thiophen-2-yl)ethyl]piperidin-4-yl}methyl carbonate,fentanyl
+CCC(=O)[N+]([O-])(c1ccccc1)C1(C(=O)OC)CCN(CCc2ccccc2)CC1,Carfentanil N-oxide,fentanyl
+CCC(=O)N(c1ccccc1)C1(C(=O)OC)CCN(CCc2cccc(O)c2)CC1,3″-Hydroxycarfentanil,fentanyl
+CCC(=O)N(c1ccccc1)C1CCN(C(C)C(=O)OC(C)C)CC1,Propan-2-yl 2-{4-[phenyl(propanoyl)amino]piperidin-1-yl}propanoate,fentanyl
+O=C(C1CCC1)N(c1ccc(Cl)cc1)C1CCN(CCc2ccccc2)CC1,p-Chloro-cyclobutylfentanyl,fentanyl
+CCC(C)C(=O)N(c1ccccc1)C1CCN(CCc2ccccc2)CC1,alpha'-Methyl-butyrylfentanyl,fentanyl
+Oc1cc2c(cc1O)CNCC2,Norsalsolinol,neurotoxin
+CC(N)Cc1ccc(Cl)cc1,PCA,neurotoxin
+CC1NCCc2cc(O)c(O)cc21,Salsolinol,neurotoxin
+CC(N)Cc1ccc(Br)cc1,PBA,neurotoxin
+NCCc1cc(O)c(O)cc1O,6-OHDA,neurotoxin
+COc1ccc(CCN)cc1OC,DMPEA,peyote alkaloid
+CNCCc1cc(OC)c(OC)c(OC)c1,M-M,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C(C)N(C=O)CC2,N-Formyl-O-methylanhalonidine,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)[N+](C)(C)CC2,Lophotine,peyote alkaloid
+COc1cc(CCN)ccc1O,GEA,peyote alkaloid
+COc1cc(CCNCC(=O)O)cc(OC)c1OC,Mescaloxylic acid,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)N(C=O)CC2,N-Formylanhalonidine,peyote alkaloid
+COc1cc(CCN2C(=O)CC(O)C2=O)cc(OC)c1OC,Mescaline malimide,peyote alkaloid
+CNCCc1ccc(O)c(O)c1,Epinine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)(C(=O)O)NCC2,Peyoruvic acid,peyote alkaloid
+CCN1CCc2cc(OC)c3c(c2C1C)OCO3,Peyophorine,peyote alkaloid
+COc1cc(CCN2C(=O)C=CC2=O)cc(OC)c1OC,Mescaline maleimide,peyote alkaloid
+COc1c(O)cc2c(c1OC)CN(C)CC2,Isoanhalidine,peyote alkaloid
+NCCc1ccc2c(c1)OCO2,MDPEA,peyote alkaloid
+COc1cc2c(c(O)c1OC)C1CCC(=O)N1CC2,Peyoglutam,peyote alkaloid
+COc1cc(CCN)cc(O)c1OC,3-DESMETHYL,peyote alkaloid
+CCOc1ccc(CC(C)N(C)C)c(OCC)c1,"2,4-Diethoxy-N,N-dimethylamphetamine",peyote alkaloid
+CN(C)CCc1ccc2c(c1)OCO2,Lobivine,peyote alkaloid
+COc1c(O)cc2c(c1OC)C(C)N(C)CC2,Isopellotine,peyote alkaloid
+COc1cc(CCNC(C)=O)cc(OC)c1OC,N-Acetylmescaline,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)N(C(C)=O)CC2,N-Acetylanhalonine,peyote alkaloid
+COc1cc(CCN(C)C)cc(O)c1OC,"N,N-Me-3-DESMETHYL",peyote alkaloid
+COc1c(O)cc2c(c1OC)C(C)NCC2,Isoanhalonidine,peyote alkaloid
+COc1cc(CCn2cccc2C(=O)O)cc(OC)c1OC,Peyonine,peyote alkaloid
+COc1cc2c(c(O)c1OC)CN(C(C)=O)CC2,N-Acetylanhalamine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C(=O)O)NCC2,Peyoxylic acid,peyote alkaloid
+COc1cc(CCN(C)C)ccc1O,MM-GEA,peyote alkaloid
+COc1cc2c(c(OC)c1OC)CNCC2,Anhalinine,peyote alkaloid
+CNCCc1ccc(O)cc1,HMePEA,peyote alkaloid
+COc1cc2c(c(O)c1OC)C[N+](C)(C)CC2,Anhalotine,peyote alkaloid
+COc1c(O)cc2c(c1OC)CNCC2,Isoanhalamine,peyote alkaloid
+COc1cc(CCN)cc(O)c1O,"3,4-DESMETHYL",peyote alkaloid
+COc1cc2c(c(O)c1OC)CN(C=O)CC2,N-Formylanhalamine,peyote alkaloid
+COc1cc(CCNC(C)C(=O)O)cc(OC)c1OC,Mescaloruvic acid,peyote alkaloid
+C[N+](C)(C)CCc1ccc(O)cc1,Candicine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)NCC2,Anhalonidine,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C1CCC(=O)N1CC2,Mescalotam,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)NCC2,Anhalonine,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)N(C)CC2,Pellotine,peyote alkaloid
+COc1cc2c(c(O)c1OC)CN(C)CC2,Anhalidine,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)N(C)CC2,Lophophorine,peyote alkaloid
+COc1cc(CCN)cc2c1OCO2,LOPHOPHINE,peyote alkaloid
+COc1cc(CCNC(C)=O)cc(O)c1OC,N-Acetyl-3-demethymescaline,peyote alkaloid
+COc1cc(CCN)cc(OC)c1OC,M,peyote alkaloid
+COc1cc(CCN2C(=O)CCC2=O)cc(OC)c1OC,Mescaline succinimide,peyote alkaloid
+COc1cc2c(c(O)c1OC)C(C)[N+](C)(C)CC2,Peyotine,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C(C)NCC2,O-Methylanhalonidine,peyote alkaloid
+COc1cc2c(c(O)c1OC)CNCC2,Anhalamine,peyote alkaloid
+COc1cc(CCN2C(=O)CC(O)(CC(=O)O)C2=O)cc(OC)c1OC,Mescaline citrimide,peyote alkaloid
+CN(C)CCc1ccc(O)cc1,Hordenine,peyote alkaloid
+COc1cc(CCNC=O)cc(OC)c1OC,N-Formylmescaline,peyote alkaloid
+CNCCc1cc(O)c(OC)c(OC)c1,N-Me-3-DESMETHYL,peyote alkaloid
+COc1cc(CCn2c(C=O)ccc2CO)cc(OC)c1OC,Peyoglunal,peyote alkaloid
+COc1cc(CCNC=O)cc(O)c1OC,N-Formyl-3-demethylmescaline,peyote alkaloid
+NCCc1ccc(O)c(O)c1,Dopamine,peyote alkaloid
+COc1cc2c(c3c1OCO3)C(C)N(C=O)CC2,N-Formylanhalonine,peyote alkaloid
+CNCCc1ccc(O)c(OC)c1,N-Me-GEA,peyote alkaloid
+COc1cc2c(c(OC)c1OC)CN(C=O)CC2,N-Formylanhalinine,peyote alkaloid
+COc1cc(CCN2C(=O)C3CC(=O)OC3C2=O)cc(OC)c1OC,Mescaline isocitrimide lactone,peyote alkaloid
+COc1cc2c(c(OC)c1OC)C(C)N(C)CC2,O-Methylpellotine,peyote alkaloid
+CCC(C)N(CCc1c[nH]c2ccc(F)cc12)C(C)C,5-Fluoro-IPSBT,tryptamine
+CC(C)N(CCc1c[nH]c2ccc(F)cc12)CC1CC1,5-Fluoro-N-cyclopropylmethyl-NIPT,tryptamine
+CCCCCCN(C)CCc1c[nH]c2ccc(OC)cc12,N-[2-(5-Methoxy-1H-indol-3-yl)ethyl]-N-methylhexan-1-amine,tryptamine
+CN(C)CCc1c[nH]c2c(Br)ccc(O)c12,7-Bromopsilocin,tryptamine
+CCCCN(CCCC)CCc1c[nH]c2cc(OC)c(OC)cc12,"5,6-MeO-DBT",tryptamine
+CN(C/C=C/c1ccccc1)CCc1c[nH]c2ccccc12,N-Phenylpropenyl-NMT,tryptamine
+COc1cc2[nH]cc(CCN3CCCC3)c2cc1OC,"5,6-MeO-pyr-T",tryptamine
+Cn1cc(CCN)c2c3c(ccc21)OCCC3,"CP 132,484",tryptamine
+C=CCN(CCc1c(C)[nH]c2ccc(OC)cc12)C1CCCCC1,2-Methyl-5-methoxy-N-allyl-N-cyclohexyltryptamine,tryptamine
+Cc1cc(OP(=O)(O)O)c2c(CCN)c[nH]c2c1,6-Methyl-4-PO-T,tryptamine
+CCCCCN(CCCCC)CCc1c[nH]c2ccccc12,"N,N-Dipentyltryptamine",tryptamine
+COc1ccc2[nH]cc(CC3CCCN3C)c2c1,CP-108509,tryptamine
+COc1ccc(OC)c(CCCN(C)CCc2c[nH]c3ccc(OC)cc23)c1,"3-(2,5-Dimethoxyphenyl)-N-[2-(5-methoxy-1H-indol-3-yl)ethyl]-N-methylpropan-1-amine",tryptamine
+C=CCN(CCc1c[nH]c2ccc(F)cc12)C(C)C,5-Fluoro-ALIPT,tryptamine
+NCCc1c[nH]c2ccc(OC3CCCC3)cc12,5-Cyclopentyloxy-T,tryptamine
+COc1ccc(OC)c(CCN(C)CCc2c[nH]c3ccc(OC)cc23)c1,"N-(2,5-Dimethoxyphenylethyl)-5-MeO-NMT",tryptamine
+Oc1cccc2[nH]cc(CCN3CCCCC3)c12,4-HO-pip-T,tryptamine
+CN(C)CCc1c[nH]c2cccc(OCc3ccccc3)c12,4-Benzyloxy-DMT,tryptamine
+CN(C)CCc1c[nH]c2cccc(OC(=O)SCc3ccccc3)c12,S-Benzyl O-{3-[2-(dimethylamino)ethyl]-1H-indol-4-yl} carbonothioate,tryptamine
+COc1cc(OC)c2[nH]cc(CCNCc3ccccc3)c2c1,"5,7-MeO-T-NB",tryptamine
+CNC(C)Cc1c[nH]c2ccc(OCc3ccccc3)cc12,5-Benzyloxy-alpha-methyl-NMT,tryptamine
+CC(C)(C)CCCCCCCOc1ccc2[nH]cc(CCN)c2c1,"5-(8,8-Dimethylnonyloxy)-T",tryptamine
+CCCOc1ccc2[nH]cc(CCN(CC)CC)c2c1,5-PrO-DET,tryptamine
+COc1ccc2[nH]cc(CCN(C)CCCc3cc(OC)c(OC)c(OC)c3)c2c1,"N-(3,4,5-Trimethoxyphenylpropyl)-5-MeO-NMT",tryptamine
+c1ccc2c(CCN3CCN(CCc4c[nH]c5ccccc45)CC3)c[nH]c2c1,"3,3'-[Piperazine-1,4-diyldi(ethane-2,1-diyl)]di(1H-indole)",tryptamine
+CC(C)N(CCc1c[nH]c2cccc(OCc3ccccc3)c12)C(C)C,4-Benzyloxy-DIPT,tryptamine
+CC(C)N(CCc1c[nH]c2ccc(F)cc12)C1CCC1,5-Fluoro-IPCBT,tryptamine
+CCOc1ccc(CNCCc2c[nH]c3ccccc23)cc1,T-NB4OEt,tryptamine
+O=C(O)CCNCCc1c[nH]c2cc([N+](=O)[O-])ccc12,6-Nitro-N-(2-carboxyethyl)tryptamine,tryptamine
+CN1CCN(CCc2c[nH]c3cccc(O)c23)CC1,4'-Methyl-4-HO-piperazine-T,tryptamine
+CN(CCC#N)CCc1c[nH]c2ccccc12,N-Cyanoethyl-NMT,tryptamine
+COc1cc(F)c2[nH]cc(CCN(C)C)c2c1,7-Fluoro-5-MeO-DMT,tryptamine
+CN(CCc1c[nH]c2ccc(F)cc12)C1CCC1,5-Fluoro-MCBT,tryptamine
+CNCCc1c[nH]c2c(OC)cccc12,7-MeO-NMT,tryptamine
+Cc1c(O)c(O)cc2c(CCN)c[nH]c12,"7-Methyl-5,6-HO-T",tryptamine
+Oc1ccc(CNCCc2c[nH]c3ccccc23)cc1,T-NB4OH,tryptamine
+COc1c(Br)cc(Br)cc1CNCCc1c[nH]c2ccccc12,T-NBOMe35Br,tryptamine
+CCn1cc(CCN(C)C)c2c(OCc3ccccc3)cc(F)cc21,1-Ethyl-6-fluoro-4-benzyloxy-DMT,tryptamine
+CN(C)CCc1c[nH]c2cccc(OC(=O)c3cccc(C(=O)Oc4cccc5[nH]cc(CCN(C)C)c45)c3)c12,"Bis{3-[2-(dimethylamino)ethyl]-1H-indol-4-yl} benzene-1,3-dicarboxylate",tryptamine
+CCN(CC)C(=O)CCN(C)CCc1c[nH]c2ccc(OC)cc12,"N,N-Diethyl-N3-[2-(5-methoxy-1H-indol-3-yl)ethyl]-N3-methyl-beta-alaninamide",tryptamine
+COc1ccc(CNCCc2c[nH]c3c(Br)cc(OC)cc23)cc1,7-Bromo-5-MeO-T-NB4OMe,tryptamine
+CCCN(C)CCc1c[nH]c2ccc(F)cc12,5-Fluoro-MPT,tryptamine
+CN(CCc1c[nH]c2cc3c(cc12)OCO3)Cc1ccccc1,"5,6-MDO-NMT-NB",tryptamine
+Cc1c(OC(=O)CCC(=O)O)ccc2[nH]cc(CCN(C)C)c12,Psilocin-4-methyl-5-succinate,tryptamine
+COc1ccc2[nH]cc(CC(C)N)c2c1Br,4-Bromo-5-MeO-AMT,tryptamine
+CN(C)CC(=O)c1c[nH]c2cc(F)ccc12,beta-Oxo-6-fluoro-DMT,tryptamine
+COc1ccc2[nH]cc(C(O)CN(C)C)c2c1,beta-HO-5-MeO-DMT,tryptamine
+COc1cccc2c(CCN)c[nH]c12,7-MeO-T,tryptamine
+COc1cc(I)c2[nH]cc(CCN(C)C)c2c1,7-Iodo-5-MeO-DMT,tryptamine
+C[C@@H]1CC[C@H](C)N1CCc1c[nH]c2ccccc12,"(R,S)-Dimethyl-pyr-T",tryptamine
+COc1ccc(Sn2cc(CCN(C)C)c3c(OC)cccc32)cc1,1-(4-Methoxyphenylthio)-4-MeO-DMT,tryptamine
+CCCN(CCc1c[nH]c2c(F)cccc12)C1CCC1,7-Fluoro-PCBT,tryptamine
+CC(=O)NCCc1c[nH]c2cc(F)cc(NC(C)=O)c12,6-Fluoro-4-acetamido-N-acetyltryptamine,tryptamine
+CCCCS(=O)(=O)n1cc(CCN(C)C)c2c(OC)cccc21,1-Butylsulfonyl-4-MeO-DMT,tryptamine
+CCCN(CCC)CCc1c[nH]c2ccc(OC(C)=O)cc12,5-AcO-DPT,tryptamine
+CN(C)CCc1cn(C)c2ccccc12,1-TMT,tryptamine
+CCN(CCc1c[nH]c2cc(F)c(F)cc12)C(C)C,"5,6-Difluoro-EIPT",tryptamine
+CN(C)CC(=O)c1c[nH]c2c(F)cc(F)cc12,"beta-Oxo-5,7-difluoro-DMT",tryptamine
+CN(C)CCc1cn(C)c2cc(F)cc(OCc3ccccc3)c12,1-Methyl-6-fluoro-4-benzyloxy-DMT,tryptamine
+CCCc1[nH]c2ccc(OC)cc2c1CCN(C)C,2-Propyl-5-MeO-DMT,tryptamine
+CC(=O)Oc2cccc1[nH]cc(CCN(C)C)c12,4-AcO-DMT,tryptamine

src/src/data/sample_smiles.csv

@@ -0,0 +1,13 @@
+smiles,name,category
+CCO,Ethanol,Alcohol
+CC(=O)O,Acetic acid,Carboxylic acid
+c1ccccc1,Benzene,Aromatic hydrocarbon
+CC(C)O,Isopropanol,Alcohol
+c1ccc(cc1)O,Phenol,Aromatic alcohol
+CC(=O)OC1=CC=CC=C1C(=O)O,Aspirin,Carboxylic acid ester
+CN1C=NC2=C1C(=O)N(C(=O)N2C)C,Caffeine,Alkaloid
+CCN(CC)CC,Triethylamine,Amine
+c1ccc2c(c1)cccn2,Quinoline,Aromatic heterocycle
+c1ccc(cc1)N,Aniline,Aromatic amine
+O=P(O)(O)OP(=O)(O)OP(=O)(O)OC[C@H]3O[C@@H](n2cnc1c(ncnc12)N)[C@H](O)[C@@H]3O,ATP,Nucleotide
+CN(C)CCC1=CNC2=C1C(=CC=C2)OP(=O)(O)O,Psilocybin,Tryptamine

src/src/main.js

@@ -0,0 +1,322 @@
+/**
+ * @fileoverview JSME (JavaScript Molecule Editor) integration with Gradio interface
+ * Handles bidirectional synchronization between JSME applet and Gradio textbox
+ * @author Manny Cortes ('[email protected]')
+ * @version 0.2.0
+ */
+
+// ============================================================================
+// GLOBAL VARIABLES
+// ============================================================================
+
+/** @type {Object|null} The JSME applet instance */
+let jsmeApplet = null;
+
+/** @type {string} Last known value of the textbox to prevent infinite loops */
+let lastTextboxValue = "";
+
+// ============================================================================
+// CONSTANTS
+// ============================================================================
+
+/** @const {string} Default SMILES for initial molecule (ethanol) */
+const DEFAULT_SMILES = "CCO";
+
+/** @const {string} Container height for JSME applet */
+const CONTAINER_HEIGHT = "450px";
+
+/** @const {string} CSS selector for the Gradio SMILES input element */
+const SMILES_INPUT_SELECTOR = "#smiles_input textarea, #smiles_input input";
+
+/** @const {number} Delay for paste event handling (ms) */
+const PASTE_DELAY = 50;
+
+/** @const {number} Delay for initialization retry (ms) */
+const INIT_RETRY_DELAY = 2000;
+
+/** @const {string[]} Events to trigger for Gradio change detection */
+const GRADIO_CHANGE_EVENTS = ["input", "change", "keyup"];
+
+// ============================================================================
+// CORE INITIALIZATION
+// ============================================================================
+
+/**
+ * Initializes the JSME applet after the library has been loaded
+ * Sets up the molecular editor with default options and callbacks
+ * @throws {Error} When JSME initialization fails
+ */
+function initializeJSME() {
+  try {
+    console.log("Initializing JSME...");
+    // https://github.com/jsme-editor/jsme-editor.github.io
+    // https://jsme-editor.github.io/dist/api_javadoc/index.html
+    // http://wiki.jmol.org/index.php/Jmol_JavaScript_Object/JME/Options
+    jsmeApplet = new JSApplet.JSME(
+      "jsme_container",
+      getJsmeContainerWidthPx(),
+      CONTAINER_HEIGHT,
+      {
+        options:
+          "rButton,zoom,zoomgui,newLook,star,multipart,polarnitro,NOexportInChI,NOexportInChIkey,NOsearchInChIkey,NOexportSVG,NOpaste",
+      }
+    );
+
+    jsmeApplet.setCallBack("AfterStructureModified", handleJSMEStructureChange);
+    jsmeApplet.setMenuScale(getJsmeGuiScale());
+    jsmeApplet.setUserInterfaceBackgroundColor("#adadad");
+
+    // Set initial molecule and sync state
+    jsmeApplet.readGenericMolecularInput(DEFAULT_SMILES);
+    lastTextboxValue = DEFAULT_SMILES;
+
+    setupTextboxEventListeners();
+    window.addEventListener("resize", handleResize);
+
+    console.log("JSME initialized successfully");
+  } catch (error) {
+    console.error("Error initializing JSME:", error);
+    throw error;
+  }
+}
+
+/**
+ * Handles structure changes in the JSME applet
+ * Converts the structure to SMILES and updates the Gradio textbox
+ * @param {Event} event - The JSME structure modification event
+ */
+function handleJSMEStructureChange(event) {
+  try {
+    const smiles = jsmeApplet.smiles();
+    updateGradioTextbox(smiles);
+  } catch (error) {
+    console.error("Error getting SMILES from JSME:", error);
+  }
+}
+
+/**
+ * Calculates the appropriate GUI scale for the JSME applet based on container width
+ * Uses breakpoints to determine optimal scaling for different screen sizes
+ * @returns {number} The scale factor for the JSME GUI (0.88 to 2.0)
+ */
+function getJsmeGuiScale() {
+  const width = getJsmeContainerWidthNumber();
+  let menuScale;
+  if (width > 460) {
+    menuScale = 1.3;
+  } else if (width > 420) {
+    menuScale = 1.1;
+  } else if (width > 370) {
+    menuScale = 1.05;
+  } else if (width > 300) {
+    menuScale = 0.88;
+  } else {
+    menuScale = 2;
+  }
+  return menuScale;
+}
+
+/**
+ * Gets the JSME container width as a CSS-compatible string value
+ * Returns either a pixel value or percentage based on available width
+ * @returns {string} Width as "100%" or "{width}px" format
+ */
+function getJsmeContainerWidthPx() {
+  const parentWidth = getJsmeContainerWidthNumber();
+  if (parentWidth == null || parentWidth <= 0) {
+    return "100%";
+  }
+  return `${parentWidth}px`;
+}
+
+/**
+ * Gets the numeric width of the JSME container's parent element
+ * Used for responsive scaling calculations
+ * @returns {number|null} Width in pixels, or null if container not found
+ */
+function getJsmeContainerWidthNumber() {
+  const container = document.getElementById("jsme_container");
+  if (!container) {
+    return null;
+  }
+  return container.parentNode.offsetWidth;
+}
+
+// ============================================================================
+// GRADIO INTEGRATION
+// ============================================================================
+
+/**
+ * Updates the Gradio textbox with a SMILES string
+ * Triggers appropriate events to ensure Gradio detects the change
+ * @param {string} smiles - The SMILES string to set in the textbox
+ */
+function updateGradioTextbox(smiles) {
+  try {
+    const textbox = document.querySelector(SMILES_INPUT_SELECTOR);
+
+    if (!textbox || textbox.value === smiles) {
+      return;
+    }
+
+    textbox.value = smiles;
+    lastTextboxValue = smiles;
+
+    // Trigger events to ensure Gradio detects the change
+    GRADIO_CHANGE_EVENTS.forEach((eventType) => {
+      const event = new Event(eventType, {
+        bubbles: true,
+        cancelable: true,
+      });
+      textbox.dispatchEvent(event);
+    });
+  } catch (error) {
+    console.error("Error updating Gradio textbox:", error);
+  }
+}
+
+// ============================================================================
+// JSME UPDATE FUNCTIONS
+// ============================================================================
+
+/**
+ * Updates the JSME applet with a SMILES string from the textbox
+ * @param {string} smiles - The SMILES string to display in JSME
+ */
+function updateJSMEFromTextbox(smiles) {
+  if (!jsmeApplet) {
+    return;
+  }
+
+  try {
+    if (smiles && smiles.trim() !== "") {
+      jsmeApplet.readGenericMolecularInput(smiles.trim());
+    } else {
+      jsmeApplet.reset();
+    }
+    lastTextboxValue = smiles;
+  } catch (error) {
+    console.error("Error updating JSME from textbox:", error);
+  }
+}
+
+// ============================================================================
+// UI MONITORING
+// ============================================================================
+
+/**
+ * Finds the textbox element and sets up event listeners
+ */
+function setupTextboxEventListeners() {
+  const textbox = document.querySelector(SMILES_INPUT_SELECTOR);
+  if (!textbox) {
+    return;
+  }
+
+  textbox.addEventListener("input", handleTextboxChange);
+  textbox.addEventListener("change", handleTextboxChange);
+  textbox.addEventListener("paste", handleTextboxPaste);
+  textbox.addEventListener("keyup", handleTextboxChange);
+}
+
+/**
+ * Handles textbox change events
+ * @param {Event} event - The change event
+ */
+function handleTextboxChange(event) {
+  if (event.target.value !== lastTextboxValue) {
+    updateJSMEFromTextbox(event.target.value);
+  }
+}
+
+/**
+ * Handles textbox paste events with a delay to ensure content is available
+ * @param {Event} event - The paste event
+ */
+function handleTextboxPaste(event) {
+  setTimeout(() => {
+    updateJSMEFromTextbox(event.target.value);
+  }, PASTE_DELAY);
+}
+
+/**
+ * Handles window resize events and updates JSME applet width
+ */
+function handleResize() {
+  if (!jsmeApplet) {
+    return;
+  }
+
+  try {
+    jsmeApplet.setMenuScale(getJsmeGuiScale());
+    jsmeApplet.setWidth(getJsmeContainerWidthPx());
+  } catch (error) {
+    console.error("Error resizing JSME applet:", error);
+  }
+}
+
+// ============================================================================
+// PUBLIC API
+// ============================================================================
+
+/**
+ * Sets a SMILES string in both JSME and Gradio textbox
+ * @param {string} smiles - The SMILES string to set
+ * @returns {string} The SMILES string that was set
+ * @public
+ */
+window.setJSMESmiles = function (smiles) {
+  if (jsmeApplet) {
+    updateJSMEFromTextbox(smiles);
+  }
+
+  updateGradioTextbox(smiles);
+  return smiles;
+};
+
+/**
+ * Clears both JSME and Gradio textbox
+ * @returns {Array} Array containing cleared state for Gradio components
+ * @public
+ */
+window.clearJSME = function () {
+  if (jsmeApplet) {
+    jsmeApplet.reset();
+  }
+
+  updateGradioTextbox("");
+  return ["", [], [], "Cleared - Draw a new molecule or enter SMILES"];
+};
+
+// ============================================================================
+// INITIALIZATION LOGIC
+// ============================================================================
+
+/**
+ * Checks if JSME library is loaded and initializes JSME applet
+ * Retries until the library becomes available
+ */
+function initializeWhenReady() {
+  if (typeof JSApplet !== "undefined" && JSApplet.JSME) {
+    console.log("JSME library loaded, initializing...");
+    initializeJSME();
+  } else {
+    console.log("JSME library not ready, retrying...");
+    setTimeout(initializeWhenReady, INIT_RETRY_DELAY);
+  }
+}
+
+/**
+ * Starts the initialization process based on document ready state
+ */
+function startInitialization() {
+  if (document.readyState === "loading") {
+    document.addEventListener("DOMContentLoaded", () => {
+      setTimeout(initializeWhenReady, INIT_RETRY_DELAY);
+    });
+  } else {
+    setTimeout(initializeWhenReady, INIT_RETRY_DELAY);
+  }
+}
+
+startInitialization();

src/src/main.min.js

@@ -0,0 +1 @@
+function initializeJSME(){try{(jsmeApplet=new JSApplet.JSME("jsme_container",getJsmeContainerWidthPx(),CONTAINER_HEIGHT,{options:"rButton,zoom,zoomgui,newLook,star,multipart,polarnitro,NOexportInChI,NOexportInChIkey,NOsearchInChIkey,NOexportSVG,NOpaste"})).setCallBack("AfterStructureModified",handleJSMEStructureChange),jsmeApplet.setMenuScale(getJsmeGuiScale()),jsmeApplet.setUserInterfaceBackgroundColor("#adadad"),jsmeApplet.readGenericMolecularInput(DEFAULT_SMILES),lastTextboxValue=DEFAULT_SMILES,setupTextboxEventListeners(),window.addEventListener("resize",handleResize)}catch(e){throw e}}function handleJSMEStructureChange(){try{updateGradioTextbox(jsmeApplet.smiles())}catch(e){}}function getJsmeGuiScale(){const e=getJsmeContainerWidthNumber();return e>460?1.3:e>420?1.1:e>370?1.05:e>300?.88:2}function getJsmeContainerWidthPx(){const e=getJsmeContainerWidthNumber();return null==e||e<=0?"100%":`${e}px`}function getJsmeContainerWidthNumber(){const e=document.getElementById("jsme_container");return e?e.parentNode.offsetWidth:null}function updateGradioTextbox(e){try{const t=document.querySelector(SMILES_INPUT_SELECTOR);if(!t||t.value===e)return;t.value=e,lastTextboxValue=e,GRADIO_CHANGE_EVENTS.forEach(e=>{const n=new Event(e,{bubbles:!0,cancelable:!0});t.dispatchEvent(n)})}catch(e){}}function updateJSMEFromTextbox(e){if(jsmeApplet)try{e&&""!==e.trim()?jsmeApplet.readGenericMolecularInput(e.trim()):jsmeApplet.reset(),lastTextboxValue=e}catch(e){}}function setupTextboxEventListeners(){const e=document.querySelector(SMILES_INPUT_SELECTOR);e&&(e.addEventListener("input",handleTextboxChange),e.addEventListener("change",handleTextboxChange),e.addEventListener("paste",handleTextboxPaste),e.addEventListener("keyup",handleTextboxChange))}function handleTextboxChange(e){e.target.value!==lastTextboxValue&&updateJSMEFromTextbox(e.target.value)}function handleTextboxPaste(e){setTimeout(()=>{updateJSMEFromTextbox(e.target.value)},PASTE_DELAY)}function handleResize(){if(jsmeApplet)try{jsmeApplet.setMenuScale(getJsmeGuiScale()),jsmeApplet.setWidth(getJsmeContainerWidthPx())}catch(e){}}function initializeWhenReady(){"undefined"!=typeof JSApplet&&JSApplet.JSME?initializeJSME():setTimeout(initializeWhenReady,INIT_RETRY_DELAY)}function startInitialization(){"loading"===document.readyState?document.addEventListener("DOMContentLoaded",()=>{setTimeout(initializeWhenReady,INIT_RETRY_DELAY)}):setTimeout(initializeWhenReady,INIT_RETRY_DELAY)}let jsmeApplet=null,lastTextboxValue="";const DEFAULT_SMILES="CCO",CONTAINER_HEIGHT="450px",SMILES_INPUT_SELECTOR="#smiles_input textarea, #smiles_input input",PASTE_DELAY=50,INIT_RETRY_DELAY=2e3,GRADIO_CHANGE_EVENTS=["input","change","keyup"];window.setJSMESmiles=function(e){return jsmeApplet&&updateJSMEFromTextbox(e),updateGradioTextbox(e),e},window.clearJSME=function(){return jsmeApplet&&jsmeApplet.reset(),updateGradioTextbox(""),["",[],[],"Cleared - Draw a new molecule or enter SMILES"]},startInitialization();

src/src/service.py

@@ -0,0 +1,213 @@
+import logging
+import os
+from typing import TypedDict
+
+import numpy as np
+import pandas as pd
+import redis
+from chem_mrl.molecular_fingerprinter import MorganFingerprinter
+from dotenv import load_dotenv
+from redis.commands.search.field import TextField, VectorField
+from redis.commands.search.indexDefinition import IndexDefinition, IndexType
+from redis.commands.search.query import Query
+from sentence_transformers import SentenceTransformer
+
+from constants import (
+    EMBEDDING_DIMENSION,
+    HNSW_K,
+    HNSW_PARAMETERS,
+    MODEL_NAME,
+    SUPPORTED_EMBEDDING_DIMENSIONS,
+)
+from data import DATASET_SMILES, ISOMER_DESIGN_SUBSET
+
+
+def setup_logger(clear_handler=False):
+    if clear_handler:
+        for handler in logging.root.handlers[:]:
+            logging.root.removeHandler(handler)  # issue with sentence-transformer's logging handler
+    logging.basicConfig(format="%(asctime)s - %(message)s", datefmt="%Y-%m-%d %H:%M:%S", level=logging.INFO)
+    logger = logging.getLogger(__name__)
+    return logger
+
+
+load_dotenv("../.env")
+logger = setup_logger(clear_handler=True)
+
+
+class SimilarMolecule(TypedDict):
+    smiles: str
+    name: str
+    category: str
+    score: float
+
+
+class MolecularEmbeddingService:
+    def __init__(self):
+        self.model_name = MODEL_NAME
+        self.index_name = "molecule_embeddings"
+        self.model_embed_dim = EMBEDDING_DIMENSION
+
+        self.model = self._initialize_model()
+        self.redis_client = self._initialize_redis()
+        self._initialize_datastore()
+
+    def _initialize_model(self):
+        """Initialize the Hugging Face transformers model"""
+        try:
+            model = SentenceTransformer(self.model_name)  # type: ignore
+            model.eval()
+            return model
+        except Exception as e:
+            logger.error(f"Failed to load model: {e}")
+            raise
+
+    def _initialize_redis(self):
+        """Initialize Redis connection"""
+        try:
+            redis_host = os.getenv("REDIS_HOST", "localhost")
+            redis_port = int(os.getenv("REDIS_PORT", 6379))
+            redis_password = os.getenv("REDIS_PASSWORD", None)
+            logger.info(
+                f"Connecting to Redis at {redis_host}:{redis_port} with password: {'***' if redis_password else 'None'}"
+            )
+            redis_client = redis.Redis(
+                host=redis_host,
+                port=redis_port,
+                password=redis_password,
+                decode_responses=True,
+            )
+            redis_client.ping()
+            return redis_client
+        except Exception as e:
+            logger.error(f"Failed to connect to Redis: {e}")
+            raise
+
+    def _initialize_datastore(self):
+        self.__create_hnsw_index()
+        self.__populate_sample_data(DATASET_SMILES)
+        self.__populate_sample_data(ISOMER_DESIGN_SUBSET)
+
+    def __create_hnsw_index(self):
+        """Create HNSW index for molecular embeddings"""
+        try:
+            self.redis_client.ft(self.index_name).info()
+            logger.info(f"Index {self.index_name} already exists")
+            return
+        except redis.exceptions.ResponseError:
+            pass
+
+        try:
+            schema: list[TextField | VectorField] = [
+                VectorField(
+                    self.embedding_field_name(dim),
+                    "HNSW",
+                    {
+                        **HNSW_PARAMETERS,
+                        "DIM": dim,
+                    },
+                )
+                for dim in SUPPORTED_EMBEDDING_DIMENSIONS
+            ]
+            schema.insert(0, TextField("smiles"))
+
+            self.redis_client.ft(self.index_name).create_index(
+                schema,
+                definition=IndexDefinition(prefix=[self.molecule_index_prefix("")], index_type=IndexType.HASH),
+            )
+
+            logger.info(f"Created HNSW index: {self.index_name}")
+
+        except Exception as e:
+            logger.error(f"Failed to create HNSW index: {e}")
+            raise
+
+    def __populate_sample_data(self, df: pd.DataFrame):
+        """Populate Redis with sample molecular data"""
+        logger.info("Populating Redis with sample molecular data...")
+        for _, row in df.iterrows():
+            try:
+                key = self.molecule_index_prefix(row["smiles"])
+                if self.redis_client.exists(key):
+                    continue
+
+                mapping: dict[str, bytes | str] = {
+                    self.embedding_field_name(embed_dim): self.get_molecular_embedding(
+                        row["smiles"], embed_dim
+                    ).tobytes()
+                    for embed_dim in SUPPORTED_EMBEDDING_DIMENSIONS
+                }
+                mapping = {**mapping, **row.to_dict()}
+
+                self.redis_client.hset(
+                    key,
+                    mapping=mapping,
+                )
+
+            except Exception as e:
+                logger.error(f"Failed to process molecule {row}: {e}")
+                continue
+
+        logger.info(f"Populated {len(df)} sample molecules")
+
+    def get_molecular_embedding(self, smiles: str, embed_dim: int) -> np.ndarray:
+        """Generate molecular embedding using ChemMRL"""
+        try:
+            if embed_dim <= 0:
+                raise ValueError("embed_dim must be positive")
+
+            # Preprocess smiles similarly as training data for optimal performance
+            smiles = MorganFingerprinter.canonicalize_smiles(smiles) or smiles
+
+            embedding: np.ndarray = self.model.encode(
+                [smiles],
+                show_progress_bar=False,
+                convert_to_numpy=True,
+            )[0]
+
+            return self._truncate_and_normalize_embedding(embedding, embed_dim)
+
+        except Exception as e:
+            logger.error(f"Failed to generate embedding for {smiles}: {e}")
+            raise
+
+    def _truncate_and_normalize_embedding(self, embedding: np.ndarray, embed_dim: int) -> np.ndarray:
+        """Truncate and normalize embedding"""
+        if embed_dim < len(embedding):
+            embedding = embedding[:embed_dim]
+        norms = np.linalg.norm(embedding, ord=2, keepdims=True)
+        return embedding / np.where(norms == 0, 1, norms)
+
+    def find_similar_molecules(
+        self, query_embedding: np.ndarray, embed_dim: int, k: int = HNSW_K
+    ) -> list[SimilarMolecule]:
+        """Find k most similar molecules using HNSW"""
+        try:
+            query_vector = query_embedding.astype(np.float32).tobytes()
+            query = (
+                Query(f"*=>[KNN {k} @{self.embedding_field_name(embed_dim)} $vec AS score]")
+                .sort_by("score")
+                .return_fields("smiles", "name", "category", "score")
+                .dialect(2)
+            )
+
+            results = self.redis_client.ft(self.index_name).search(query, query_params={"vec": query_vector})
+
+            neighbors: list[SimilarMolecule] = [
+                {"smiles": doc.smiles, "name": doc.name, "category": doc.category, "score": float(doc.score)}
+                for doc in results.docs
+            ]
+
+            return neighbors
+
+        except Exception as e:
+            logger.error(f"Failed to find similar molecules: {e}")
+            return []
+
+    @staticmethod
+    def embedding_field_name(dim: int) -> str:
+        return f"embedding_{dim}"
+
+    @staticmethod
+    def molecule_index_prefix(smiles: str) -> str:
+        return f"mol:{smiles}"