Some changes to text, going to try change the upload to two files

README.md

@@ -35,4 +35,4 @@ uv run python app.py
 Run tests
 ```
 uv run pytest
-```
+```

about.py

@@ -4,6 +4,8 @@ from constants import (
     SUBMIT_TAB_NAME,
     TERMS_URL,
     FAQ_TAB_NAME,
+    SLACK_URL,
+    TUTORIAL_URL,
 )
 
 ABOUT_INTRO = f"""
@@ -21,7 +23,7 @@ Here we invite the community to submit and develop better predictors, which will
 
 For each of the 5 properties in the competition, there is a prize for the model with the highest performance for that property on the private test set.
 There is also an 'open-source' prize for the best model trained on the GDPa1 dataset of monoclonal antibodies (reporting cross-validation results) and assessed on the private test set where authors provide all training code and data.
-For each of these 6 prizes, participants have the choice between **$10k in data generation credits** with [Ginkgo Datapoints](https://datapoints.ginkgo.bio/) or a **cash prize** with a value of $2000.
+For each of these 6 prizes, participants have the choice between **$10k in data generation credits** with [Ginkgo Datapoints](https://datapoints.ginkgo.bio/) or a **cash prize** with a value of **$2000**.
 
 See the "{FAQ_TAB_NAME}" tab above (you are currently on the "{ABOUT_TAB_NAME}" tab) or the [competition terms]({TERMS_URL}) for more details.
 """
@@ -42,7 +44,7 @@ Submissions close on **1 November 2025**.
 
 #### Acknowledgements
 
-We gratefully acknowledge [Tamarind Bio](https://www.tamarind.bio/)'s help in running the following models:
+We gratefully acknowledge [Tamarind Bio](https://www.tamarind.bio/)'s help in running the following models which are on the leaderboard:
 - TAP (Therapeutic Antibody Profiler)
 - SaProt
 - DeepViscosity
@@ -54,12 +56,11 @@ We're working on getting more public models added, so that participants have mor
 #### How to contribute?
 
 We'd like to add some more existing developability models to the leaderboard. Some examples of models we'd like to add:
-- ESM embeddings + ridge regression
 - Absolute folding stability models (for Thermostability)
 - PROPERMAB
 - AbMelt (requires GROMACS for MD simulations)
 
-If you would like to collaborate with others, start a discussion on the "Community" tab at the top of this page.
+If you would like to form a team or discuss ideas, join the [Slack community]({SLACK_URL}) co-hosted by Bits in Bio.
 """
 
 # Note(Lood): Significance: Add another note of "many models are trained on different datasets, and differing train/test splits, so this is a consistent way of comparing for a heldout set"
@@ -155,8 +156,8 @@ We may release private test set results at intermediate points during the compet
 ## Cross-validation
 
 For the cross-validation metrics (if training only on the GDPa1 dataset), use the `"hierarchical_cluster_IgG_isotype_stratified_fold"` column to split the dataset into folds and make predictions for each of the folds.
-Submit a CSV file in the same format but also containing the `"hierarchical_cluster_IgG_isotype_stratified_fold"` column. 
-We will be releasing a tutorial on cross-validation shortly.
+Submit a CSV file in the same format but also containing the `"hierarchical_cluster_IgG_isotype_stratified_fold"` column.
+Check out our tutorial on making an antibody developability prediction model [here]({TUTORIAL_URL}).
 
 Submissions close on **1 November 2025**.
 """

app.py

@@ -1,10 +1,10 @@
-import hashlib
 import pandas as pd
 import gradio as gr
 from gradio.themes.utils import sizes
 from gradio_leaderboard import Leaderboard
 from dotenv import load_dotenv
 import contextlib
+
 load_dotenv()  # Load environment variables from .env file
 
 from about import ABOUT_INTRO, ABOUT_TEXT, FAQS, SUBMIT_INTRUCTIONS
@@ -18,9 +18,10 @@ from constants import (
     LEADERBOARD_COLUMNS_RENAME,
     LEADERBOARD_COLUMNS_RENAME_LIST,
     SUBMIT_TAB_NAME,
+    SLACK_URL,
 )
 from submit import make_submission
-from utils import fetch_hf_results, show_output_box, get_time
+from utils import fetch_hf_results, show_output_box
 
 
 def format_leaderboard_table(df_results: pd.DataFrame, assay: str | None = None):
@@ -71,6 +72,7 @@ def get_leaderboard_object(assay: str | None = None):
 fetch_hf_results()
 current_dataframe = pd.read_csv("debug-current-results.csv")
 
+
 def refresh_overall_leaderboard():
     current_dataframe = pd.read_csv("debug-current-results.csv")
     return format_leaderboard_table(df_results=current_dataframe)
@@ -78,6 +80,7 @@ def refresh_overall_leaderboard():
 
 def fetch_latest_data(stop_event):
     import time
+
     while not stop_event.is_set():
         try:
             fetch_hf_results()
@@ -90,6 +93,7 @@ def fetch_latest_data(stop_event):
 @contextlib.asynccontextmanager
 async def periodic_data_fetch(app):
     import threading
+
     event = threading.Event()
     t = threading.Thread(target=fetch_latest_data, args=(event,), daemon=True)
     t.start()
@@ -98,7 +102,7 @@ async def periodic_data_fetch(app):
     t.join(3)
 
 
-# Lood: Two problems currently: 
+# Lood: Two problems currently:
 # 1. The data_version state value isn't being incremented, it seems (even though it's triggering the dataframe change correctly)
 # 2. The global current_dataframe is being shared across all sessions
 
@@ -165,7 +169,7 @@ with gr.Blocks(theme=gr.themes.Default(text_size=sizes.text_lg)) as demo:
                 """
                 # Overall Leaderboard (filter below by property)
                 Each property has its own prize, and participants can submit models for any combination of properties.
-                
+
                 **Note**: It is trivial to overfit the public GDPa1 dataset, which results in very high Spearman correlations.
                 We would suggest training using cross-validation a limited number of times to give a better indication of the model's performance on the eventual private test set.
                 """
@@ -182,7 +186,9 @@ with gr.Blocks(theme=gr.themes.Default(text_size=sizes.text_lg)) as demo:
         with gr.TabItem(SUBMIT_TAB_NAME, elem_id="boundary-benchmark-tab-table"):
             gr.Markdown(SUBMIT_INTRUCTIONS)
             submission_type_state = gr.State(value="GDPa1_cross_validation")
-            download_file_state = gr.State(value=EXAMPLE_FILE_DICT["GDPa1_cross_validation"])
+            download_file_state = gr.State(
+                value=EXAMPLE_FILE_DICT["GDPa1_cross_validation"]
+            )
 
             with gr.Row():
                 with gr.Column():
@@ -215,13 +221,11 @@ with gr.Blocks(theme=gr.themes.Default(text_size=sizes.text_lg)) as demo:
                         placeholder="Enter your registration code",
                         info="If you did not receive a registration code, please sign up on the <a href='https://datapoints.ginkgo.bio/ai-competitions/2025-abdev-competition'>Competition Registration page</a> or email <a href='mailto:[email protected]'>[email protected]</a>.",
                     )
-                    
+
                     # Extra validation / warning
                     # Add the conditional warning checkbox
                     high_corr_warning = gr.Markdown(
-                        value="",
-                        visible=False,
-                        elem_classes=["warning-box"]
+                        value="", visible=False, elem_classes=["warning-box"]
                     )
                     high_corr_checkbox = gr.Checkbox(
                         label="I understand this may be overfitting",
@@ -229,7 +233,7 @@ with gr.Blocks(theme=gr.themes.Default(text_size=sizes.text_lg)) as demo:
                         visible=False,
                         info="This checkbox will appear if your submission shows suspiciously high correlations (>0.9).",
                     )
-                    
+
                 with gr.Column():
                     submission_type_dropdown = gr.Dropdown(
                         choices=["GDPa1", "GDPa1_cross_validation", "Heldout Test Set"],
@@ -275,10 +279,6 @@ with gr.Blocks(theme=gr.themes.Default(text_size=sizes.text_lg)) as demo:
 
             submit_btn = gr.Button("Evaluate")
             message = gr.Textbox(label="Status", lines=1, visible=False)
-            # help message
-            gr.Markdown(
-                "If you have issues with submission or using the leaderboard, please start a discussion in the Community tab of this Space."
-            )
 
             submit_btn.click(
                 make_submission,
@@ -309,12 +309,14 @@ with gr.Blocks(theme=gr.themes.Default(text_size=sizes.text_lg)) as demo:
     gr.Markdown(
         f"""
         <div style="text-align: center; font-size: 14px; color: gray; margin-top: 2em;">
-        📬 For questions or feedback, contact <a href="mailto:[email protected]">[email protected]</a> or visit the Community tab at the top of this page.<br>
-        Visit the <a href="https://datapoints.ginkgo.bio/ai-competitions/2025-abdev-competition">Competition Registration page</a> to sign up for updates and to register a team, and see Terms <a href="{TERMS_URL}">here</a>.
+        📬 For questions or feedback, contact <a href="mailto:[email protected]">[email protected]</a> or discuss on the <a href="{SLACK_URL}">Slack community</a> co-hosted by Bits in Bio.<br>
+        Visit the <a href="https://datapoints.ginkgo.bio/ai-competitions/2025-abdev-competition">Competition Registration page</a> to sign up for updates and to register, and see Terms <a href="{TERMS_URL}">here</a>.
         </div>
         """,
         elem_id="contact-footer",
     )
 
 if __name__ == "__main__":
-    demo.launch(ssr_mode=False, share=False, app_kwargs={"lifespan": periodic_data_fetch})
+    demo.launch(
+        ssr_mode=False, share=True, app_kwargs={"lifespan": periodic_data_fetch}
+    )

constants.py

@@ -4,7 +4,6 @@ Constants for the Antibody Developability Benchmark
 
 import os
 from huggingface_hub import HfApi
-import pandas as pd
 
 ASSAY_LIST = ["AC-SINS_pH7.4", "PR_CHO", "HIC", "Tm2", "Titer"]
 ASSAY_RENAME = {
@@ -42,6 +41,8 @@ SUBMIT_TAB_NAME = "✉️ Submit"
 
 REGISTRATION_CODE = os.environ.get("REGISTRATION_CODE")
 TERMS_URL = "https://euphsfcyogalqiqsawbo.supabase.co/storage/v1/object/public/gdpweb/pdfs/2025%20Ginkgo%20Antibody%20Developability%20Prediction%20Competition%202025-08-28-v2.pdf"
+SLACK_URL = "https://join.slack.com/t/bitsinbio/shared_invite/zt-3dqigle2b-e0dEkfPPzzWL055j_8N_eQ"
+TUTORIAL_URL = "https://huggingface.co/blog/ginkgo-datapoints/making-antibody-embeddings-and-predictions"
 
 # Input CSV file requirements
 REQUIRED_COLUMNS: list[str] = [

data/example-predictions-cv.csv

@@ -244,4 +244,4 @@ visilizumab,QVQLVQSGAEVKKPGASVKVSCKASGYTFISYTMHWVRQAPGQGLEWMGYINPRSGYTHYNQKLKDKA
 xentuzumab,QVELVESGGGLVQPGGSLRLSCAASGFTFTSYWMSWVRQAPGKGLELVSSITSYGSFTYYADSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARNMYTHFDSWGQGTLVTVSS,DIVLTQPPSVSGAPGQRVTISCSGSSSNIGSNSVSWYQQLPGTAPKLLIYDNSKRPSGVPDRFSGSKSGTSASLAITGLQSEDEADYYCQSRDTYGYYWVFGGGTKLTVL,4,-0.6543
 zalutumumab,QVQLVESGGGVVQPGRSLRLSCAASGFTFSTYGMHWVRQAPGKGLEWVAVIWDDGSYKYYGDSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDGITMVRGVMKDYFDYWGQGTLVTVSS,AIQLTQSPSSLSASVGDRVTITCRASQDISSALVWYQQKPGKAPKLLIYDASSLESGVPSRFSGSESGTDFTLTISSLQPEDFATYYCQQFNSYPLTFGGGTKVEIK,0,-0.5345
 zanolimumab,QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEINHSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARVINWFDPWGQGTLVTVSS,DIQMTQSPSSVSASVGDRVTITCRASQDISSWLAWYQHKPGKAPKLLIYAASSLQSGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQANSFPYTFGQGTKLEIK,3,-0.2446
-zolbetuximab,QVQLQQPGAELVRPGASVKLSCKASGYTFTSYWINWVKQRPGQGLEWIGNIYPSDSYTNYNQKFKDKATLTVDKSSSTAYMQLSSPTSEDSAVYYCTRSWRGNSFDYWGQGTTLTVSS,DIVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYQQKPGQPPKLLIYWASTRESGVPDRFTGSGSGTDFTLTISSVQAEDLAVYYCQNDYSYPFTFGSGTKLEIK,4,-0.3497
+zolbetuximab,QVQLQQPGAELVRPGASVKLSCKASGYTFTSYWINWVKQRPGQGLEWIGNIYPSDSYTNYNQKFKDKATLTVDKSSSTAYMQLSSPTSEDSAVYYCTRSWRGNSFDYWGQGTTLTVSS,DIVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYQQKPGQPPKLLIYWASTRESGVPDRFTGSGSGTDFTLTISSVQAEDLAVYYCQNDYSYPFTFGSGTKLEIK,4,-0.3497

data/example-predictions.csv

@@ -244,4 +244,4 @@ visilizumab,QVQLVQSGAEVKKPGASVKVSCKASGYTFISYTMHWVRQAPGQGLEWMGYINPRSGYTHYNQKLKDKA
 xentuzumab,QVELVESGGGLVQPGGSLRLSCAASGFTFTSYWMSWVRQAPGKGLELVSSITSYGSFTYYADSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARNMYTHFDSWGQGTLVTVSS,DIVLTQPPSVSGAPGQRVTISCSGSSSNIGSNSVSWYQQLPGTAPKLLIYDNSKRPSGVPDRFSGSKSGTSASLAITGLQSEDEADYYCQSRDTYGYYWVFGGGTKLTVL,-0.6543
 zalutumumab,QVQLVESGGGVVQPGRSLRLSCAASGFTFSTYGMHWVRQAPGKGLEWVAVIWDDGSYKYYGDSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDGITMVRGVMKDYFDYWGQGTLVTVSS,AIQLTQSPSSLSASVGDRVTITCRASQDISSALVWYQQKPGKAPKLLIYDASSLESGVPSRFSGSESGTDFTLTISSLQPEDFATYYCQQFNSYPLTFGGGTKVEIK,-0.5345
 zanolimumab,QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEINHSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARVINWFDPWGQGTLVTVSS,DIQMTQSPSSVSASVGDRVTITCRASQDISSWLAWYQHKPGKAPKLLIYAASSLQSGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQANSFPYTFGQGTKLEIK,-0.2446
-zolbetuximab,QVQLQQPGAELVRPGASVKLSCKASGYTFTSYWINWVKQRPGQGLEWIGNIYPSDSYTNYNQKFKDKATLTVDKSSSTAYMQLSSPTSEDSAVYYCTRSWRGNSFDYWGQGTTLTVSS,DIVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYQQKPGQPPKLLIYWASTRESGVPDRFTGSGSGTDFTLTISSVQAEDLAVYYCQNDYSYPFTFGSGTKLEIK,-0.3497
+zolbetuximab,QVQLQQPGAELVRPGASVKLSCKASGYTFTSYWINWVKQRPGQGLEWIGNIYPSDSYTNYNQKFKDKATLTVDKSSSTAYMQLSSPTSEDSAVYYCTRSWRGNSFDYWGQGTTLTVSS,DIVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYQQKPGQPPKLLIYWASTRESGVPDRFTGSGSGTDFTLTISSVQAEDLAVYYCQNDYSYPFTFGSGTKLEIK,-0.3497

evaluation.py

@@ -123,9 +123,7 @@ def evaluate(predictions_df, target_df, dataset_name="GDPa1"):
     eg. my_model.csv has columns antibody_name, HIC, Tm2
     Lood: Copied from Github repo, which I should move over here
     """
-    properties_in_preds = [
-        col for col in predictions_df.columns if col in ASSAY_LIST
-    ]
+    properties_in_preds = [col for col in predictions_df.columns if col in ASSAY_LIST]
     df_merged = pd.merge(
         target_df[["antibody_name", FOLD_COL] + ASSAY_LIST],
         predictions_df[["antibody_name"] + properties_in_preds],
@@ -137,15 +135,11 @@ def evaluate(predictions_df, target_df, dataset_name="GDPa1"):
     # Process each property one by one for better error handling
     for assay_col in properties_in_preds:
         try:
-            results = _get_result_for_assay(
-                df_merged, assay_col, dataset_name
-            )
+            results = _get_result_for_assay(df_merged, assay_col, dataset_name)
             results_list.append(results)
 
         except Exception as e:
-            error_result = _get_error_result(
-                assay_col, dataset_name, e
-            )
+            error_result = _get_error_result(assay_col, dataset_name, e)
             results_list.append(error_result)
 
     results_df = pd.DataFrame(results_list)

requirements.txt

@@ -1,5 +1,5 @@
-gradio
 datasets
+dotenv
 huggingface_hub
 gradio-leaderboard
 gradio[oauth]

submit.py

@@ -68,7 +68,6 @@ def make_submission(
     registration_code: str = "",
     # profile: gr.OAuthProfile | None = None,
 ):
-
     # if profile:
     #    user_state = profile.name
     # user_state = user_state
@@ -98,8 +97,6 @@ def make_submission(
     if path_obj.suffix.lower() != ".csv":
         raise gr.Error("File must be a CSV file. Please upload a .csv file.")
 
-    
-
     upload_submission(
         file_path=path_obj,
         user_state=user_state,

utils.py

@@ -1,10 +1,18 @@
 from datetime import datetime, timezone, timedelta
-import pandas as pd
-from datasets import load_dataset
-import gradio as gr
 import hashlib
+import os
 from typing import Iterable, Union
-from constants import RESULTS_REPO, ASSAY_RENAME, LEADERBOARD_RESULTS_COLUMNS, BASELINE_USERNAMES
+
+from datasets import load_dataset
+import gradio as gr
+import pandas as pd
+
+from constants import (
+    RESULTS_REPO,
+    ASSAY_RENAME,
+    LEADERBOARD_RESULTS_COLUMNS,
+    BASELINE_USERNAMES,
+)
 
 pd.set_option("display.max_columns", None)
 
@@ -15,7 +23,11 @@ def get_time(tz_name="EST") -> str:
         print("Invalid timezone, using EST")
         tz_name = "EST"
     offset = offsets[tz_name]
-    return datetime.now(timezone(timedelta(hours=offset))).strftime("%Y-%m-%d %H:%M:%S") + f" ({tz_name})"
+    return (
+        datetime.now(timezone(timedelta(hours=offset))).strftime("%Y-%m-%d %H:%M:%S")
+        + f" ({tz_name})"
+    )
+
 
 def show_output_box(message):
     return gr.update(value=message, visible=True)
@@ -32,56 +44,183 @@ def fetch_hf_results():
         RESULTS_REPO,
         data_files="auto_submissions/metrics_all.csv",
     )["train"].to_pandas()
-    print("fetched results from HF", df.shape)
     assert all(
         col in df.columns for col in LEADERBOARD_RESULTS_COLUMNS
     ), f"Expected columns {LEADERBOARD_RESULTS_COLUMNS} not found in {df.columns}. Missing columns: {set(LEADERBOARD_RESULTS_COLUMNS) - set(df.columns)}"
-    
+
     df_baseline = df[df["user"].isin(BASELINE_USERNAMES)]
     df_non_baseline = df[~df["user"].isin(BASELINE_USERNAMES)]
     # Show latest submission only
     # For baselines: Keep unique model names
-    df_baseline = df_baseline.sort_values("submission_time", ascending=False).drop_duplicates(
-        subset=["model", "assay", "dataset", "user"], keep="first"
-    )
+    df_baseline = df_baseline.sort_values(
+        "submission_time", ascending=False
+    ).drop_duplicates(subset=["model", "assay", "dataset", "user"], keep="first")
     # For users: Just show latest submission
-    df_non_baseline = df_non_baseline.sort_values("submission_time", ascending=False).drop_duplicates(
-        subset=["assay", "dataset", "user"], keep="first"
-    )
+    df_non_baseline = df_non_baseline.sort_values(
+        "submission_time", ascending=False
+    ).drop_duplicates(subset=["assay", "dataset", "user"], keep="first")
     df = pd.concat([df_baseline, df_non_baseline], ignore_index=True)
     df["property"] = df["assay"].map(ASSAY_RENAME)
-    
+
     # Rename baseline username to just "Baseline"
     df.loc[df["user"].isin(BASELINE_USERNAMES), "user"] = "Baseline"
     # Note: Could optionally add a column "is_baseline" to the dataframe to indicate whether the model is a baseline model or not. If things get crowded.
     # Anonymize the user column at this point (so note: users can submit anonymous / non-anonymous and we'll show their latest submission regardless)
-    df.loc[df["anonymous"] != False, "user"] = "anon-" + df.loc[df["anonymous"] != False, "user"].apply(readable_hash)
+    df.loc[df["anonymous"] != False, "user"] = "anon-" + df.loc[
+        df["anonymous"] != False, "user"
+    ].apply(readable_hash)
+    
+    # Compare to previous dataframe
+    if os.path.exists("debug-current-results.csv"):
+        old_df = pd.read_csv("debug-current-results.csv")
+    else:
+        old_df = df
+    if len(df) != len(old_df):
+        print(f"New results: Length {len(old_df)} -> {len(df)} ({get_time()})")
+    
     df.to_csv("debug-current-results.csv", index=False)
 
 
 # Readable hashing function similar to coolname or codenamize
 ADJECTIVES = [
-    "ancient","brave","calm","clever","crimson","curious","dapper","eager",
-    "fuzzy","gentle","glowing","golden","happy","icy","jolly","lucky",
-    "magical","mellow","nimble","peachy","quick","royal","shiny","silent",
-    "sly","sparkly","spicy","spry","sturdy","sunny","swift","tiny","vivid",
-    "witty"
+    "ancient",
+    "brave",
+    "calm",
+    "clever",
+    "crimson",
+    "curious",
+    "dapper",
+    "eager",
+    "fuzzy",
+    "gentle",
+    "glowing",
+    "golden",
+    "happy",
+    "icy",
+    "jolly",
+    "lucky",
+    "magical",
+    "mellow",
+    "nimble",
+    "peachy",
+    "quick",
+    "royal",
+    "shiny",
+    "silent",
+    "sly",
+    "sparkly",
+    "spicy",
+    "spry",
+    "sturdy",
+    "sunny",
+    "swift",
+    "tiny",
+    "vivid",
+    "witty",
 ]
 
 ANIMALS = [
-    "ant","bat","bear","bee","bison","boar","bug","cat","crab","crow",
-    "deer","dog","duck","eel","elk","fox","frog","goat","gull","hare",
-    "hawk","hen","horse","ibis","kid","kiwi","koala","lamb","lark","lemur",
-    "lion","llama","loon","lynx","mole","moose","mouse","newt","otter","owl",
-    "ox","panda","pig","prawn","puma","quail","quokka","rabbit","rat","ray",
-    "robin","seal","shark","sheep","shrew","skunk","slug","snail","snake",
-    "swan","toad","trout","turtle","vole","walrus","wasp","whale","wolf",
-    "worm","yak","zebra"
+    "ant",
+    "bat",
+    "bear",
+    "bee",
+    "bison",
+    "boar",
+    "bug",
+    "cat",
+    "crab",
+    "crow",
+    "deer",
+    "dog",
+    "duck",
+    "eel",
+    "elk",
+    "fox",
+    "frog",
+    "goat",
+    "gull",
+    "hare",
+    "hawk",
+    "hen",
+    "horse",
+    "ibis",
+    "kid",
+    "kiwi",
+    "koala",
+    "lamb",
+    "lark",
+    "lemur",
+    "lion",
+    "llama",
+    "loon",
+    "lynx",
+    "mole",
+    "moose",
+    "mouse",
+    "newt",
+    "otter",
+    "owl",
+    "ox",
+    "panda",
+    "pig",
+    "prawn",
+    "puma",
+    "quail",
+    "quokka",
+    "rabbit",
+    "rat",
+    "ray",
+    "robin",
+    "seal",
+    "shark",
+    "sheep",
+    "shrew",
+    "skunk",
+    "slug",
+    "snail",
+    "snake",
+    "swan",
+    "toad",
+    "trout",
+    "turtle",
+    "vole",
+    "walrus",
+    "wasp",
+    "whale",
+    "wolf",
+    "worm",
+    "yak",
+    "zebra",
 ]
 NOUNS = [
-    "rock","sand","star","tree","leaf","seed","stone","cloud","rain","snow",
-    "wind","fire","ash","dirt","mud","ice","wave","shell","dust","sun",
-    "moon","hill","lake","pond","reef","root","twig","wood"
+    "rock",
+    "sand",
+    "star",
+    "tree",
+    "leaf",
+    "seed",
+    "stone",
+    "cloud",
+    "rain",
+    "snow",
+    "wind",
+    "fire",
+    "ash",
+    "dirt",
+    "mud",
+    "ice",
+    "wave",
+    "shell",
+    "dust",
+    "sun",
+    "moon",
+    "hill",
+    "lake",
+    "pond",
+    "reef",
+    "root",
+    "twig",
+    "wood",
 ]
 
 

validation.py

@@ -138,7 +138,6 @@ def validate_cv_submission(
         raise gr.Error(
             f"❌ Fold assignments don't match canonical CV folds: {'; '.join(examples)}"
         )
-    
 
 
 def validate_full_dataset_submission(df: pd.DataFrame) -> None:
@@ -204,7 +203,7 @@ def validate_dataframe(df: pd.DataFrame, submission_type: str = "GDPa1") -> None
         raise gr.Error(
             f"❌ CSV should have only one row per antibody. Found {n_duplicates} duplicates."
         )
-    
+
     example_df = pd.read_csv(EXAMPLE_FILE_DICT[submission_type])
     # All antibody names should be recognizable
     unrecognized_antibodies = set(df["antibody_name"]) - set(
@@ -229,11 +228,13 @@ def validate_dataframe(df: pd.DataFrame, submission_type: str = "GDPa1") -> None
         validate_cv_submission(df, submission_type)
     else:  # full_dataset
         validate_full_dataset_submission(df)
-    
+
     # Check Spearman correlations on public set
     df_gdpa1 = pd.read_csv(GDPa1_path)
     if submission_type in ["GDPa1", "GDPa1_cross_validation"]:
-        results_df = evaluate(predictions_df=df, target_df=df_gdpa1, dataset_name=submission_type)
+        results_df = evaluate(
+            predictions_df=df, target_df=df_gdpa1, dataset_name=submission_type
+        )
         # Check that the Spearman correlations are not too high
         if results_df["spearman"].max() > 0.9:
             raise gr.Error(