Venkatachalam commited on
Commit
0d6c384
·
1 Parent(s): 7dac9a7
Files changed (1) hide show
  1. README.md +25 -19
README.md CHANGED
@@ -7,7 +7,7 @@ license: cc-by-4.0 # Or the specific license provided by the Muscle Ageing Cell
7
  multilinguality: monolingual
8
  pretty_name: Mouse Skeletal Muscle Aging Atlas (sn/scRNA-seq)
9
  size_categories:
10
- - 100K<n<1M # Based on ~96K cells, this category fits
11
  source_datasets:
12
  - original
13
  tags:
@@ -43,11 +43,11 @@ Skeletal muscle undergoes significant functional and structural decline with age
43
  This dataset offers an unprecedented opportunity to:
44
  * Identify **age-specific molecular signatures** within various skeletal muscle cell types (e.g., muscle stem cells, fibroblasts, immune cells).
45
  * Uncover how cellular processes like muscle regeneration, metabolism, inflammation, and cellular senescence change with age at the single-cell level.
46
- * Discover **biomarkers** or **therapeutic targets** for age-associated muscle decline.
47
  * Investigate the contribution of different cell types to the overall aging process of skeletal muscle and their interplay.
48
  * Analyze shifts in cellular composition within the muscle tissue with advancing age.
49
 
50
- This dataset is therefore a powerful resource for understanding the intricate molecular mechanisms of aging within a vital mammalian tissue, with direct implications for longevity and healthspan research.
51
 
52
  ---
53
 
@@ -113,30 +113,29 @@ Explore cellular heterogeneity, identify novel cell states, and characterize gen
113
  * Discover biomarkers or therapeutic targets for sarcopenia and other age-related muscle pathologies.
114
 
115
  ### Machine Learning
116
- * **Clustering:** Apply clustering algorithms (e.g., K-Means, Leiden) on `pca_embeddings.parquet` or `umap_embeddings.parquet` to identify distinct cell populations or sub-populations.
117
  * **Classification:** Build models to classify cell types, age groups (e.g., young vs. old mice), or other relevant phenotypes using `pca_embeddings.parquet` or `umap_embeddings.parquet` as features. `cell_metadata.parquet` provides the necessary labels.
118
  * **Regression:** Predict the biological age of a cell or donor based on gene expression or derived features.
119
  * **Dimensionality Reduction & Visualization:** Use the PCA and UMAP embeddings for generating 2D or 3D plots to visualize complex cell relationships and age-related trends.
120
  * **Feature Selection:** Identify key genes or principal components relevant to muscle aging processes.
121
 
122
- ### **Direct Download and Loading from Hugging Face Hub**
123
  This dataset is hosted on the Hugging Face Hub, allowing for easy programmatic download and loading of its component files.
124
 
125
  ```python
126
  import pandas as pd
127
  from huggingface_hub import hf_hub_download
128
  import os
129
- import anndata as ad # Needed if you download the .h5ad file
130
 
131
  # Define the Hugging Face repository ID and the local directory for downloads
132
- HF_REPO_ID = "longevity-db/your-new-mouse-muscle-dataset-name" # <<<--- REPLACE with your actual HF repo name
133
  LOCAL_DATA_DIR = "downloaded_mouse_muscle_data"
134
 
135
  os.makedirs(LOCAL_DATA_DIR, exist_ok=True)
136
  print(f"Created local download directory: {LOCAL_DATA_DIR}")
137
 
138
- # List of files to download
139
- data_files = [
140
  "expression.parquet",
141
  "gene_metadata.parquet",
142
  "cell_metadata.parquet",
@@ -144,27 +143,29 @@ data_files = [
144
  "pca_explained_variance.parquet",
145
  "umap_embeddings.parquet",
146
  "highly_variable_gene_metadata.parquet",
147
- "gene_statistics.parquet",
148
- # Add the original .h5ad if you upload it
149
- # "SKM_mouse_pp_cells2nuclei_2022-03-30.h5ad",
150
  ]
151
 
152
  # Download each file
153
  downloaded_paths = {}
154
- for file_name in data_files:
155
  try:
156
  path = hf_hub_download(repo_id=HF_REPO_ID, filename=file_name, local_dir=LOCAL_DATA_DIR)
157
  downloaded_paths[file_name] = path
158
  print(f"Downloaded {file_name} to: {path}")
159
  except Exception as e:
160
- print(f"Warning: Could not download {file_name}. It might not be in the repository. Error: {e}")
161
 
162
- # Load key Parquet files into Pandas DataFrames
163
  df_expression = pd.read_parquet(downloaded_paths["expression.parquet"])
164
  df_pca_embeddings = pd.read_parquet(downloaded_paths["pca_embeddings.parquet"])
165
  df_umap_embeddings = pd.read_parquet(downloaded_paths["umap_embeddings.parquet"])
166
  df_cell_metadata = pd.read_parquet(downloaded_paths["cell_metadata.parquet"])
167
  df_gene_metadata = pd.read_parquet(downloaded_paths["gene_metadata.parquet"])
 
 
 
 
168
 
169
  print("\n--- Data Loaded from Hugging Face Hub ---")
170
  print("Expression data shape:", df_expression.shape)
@@ -172,11 +173,14 @@ print("PCA embeddings shape:", df_pca_embeddings.shape)
172
  print("UMAP embeddings shape:", df_umap_embeddings.shape)
173
  print("Cell metadata shape:", df_cell_metadata.shape)
174
  print("Gene metadata shape:", df_gene_metadata.shape)
 
 
 
175
 
176
 
177
  # Example: Prepare data for an age prediction model
178
  # IMPORTANT: You need to inspect `df_cell_metadata.columns` to find the actual age and cell type columns.
179
- print("\nAvailable columns in cell_metadata.parquet:")
180
  print(df_cell_metadata.columns.tolist())
181
 
182
  # --- USER ACTION REQUIRED ---
@@ -194,7 +198,7 @@ if age_column_name in df_cell_metadata.columns:
194
  y_labels_age_prediction = df_cell_metadata[age_column_name]
195
  print(f"\nPrepared X (features) for age prediction with shape {X_features_age_prediction.shape} and y (labels) with shape {y_labels_age_prediction.shape}")
196
  else:
197
- print(f"\nWarning: Column '{age_column_name}' not found in cell metadata for age prediction example.")
198
 
199
  # Example: Using cell type for a classification task
200
  if cell_type_column_name in df_cell_metadata.columns:
@@ -202,7 +206,7 @@ if cell_type_column_name in df_cell_metadata.columns:
202
  y_labels_cell_type = df_cell_metadata[cell_type_column_name]
203
  print(f"Prepared X (features) for cell type classification with shape {X_features_cell_type.shape} and y (labels) with shape {y_labels_cell_type.shape}")
204
  else:
205
- print(f"Warning: Column '{cell_type_column_name}' not found in cell metadata for cell type classification example.")
206
 
207
  # This data can then be split into train/test sets and used to train various ML models.
208
  ```
@@ -214,6 +218,8 @@ Please ensure you cite the original source of the Mouse Muscle Ageing Cell Atlas
214
  **Mouse Muscle Ageing Cell Atlas Official Website:**
215
  [https://www.muscleageingcellatlas.org/mouse-pp/](https://www.muscleageingcellatlas.org/mouse-pp/)
216
 
 
 
217
  ## Contributions
218
 
219
  This dataset was processed and prepared by:
@@ -224,4 +230,4 @@ This dataset was processed and prepared by:
224
 
225
  *Curated on June 15, 2025.*
226
 
227
- **Hugging Face Repository:** [https://huggingface.co/datasets/longevity-db/your-new-mouse-muscle-dataset-name](https://www.google.com/search?q=https://huggingface.co/datasets/longevity-db/your-new-mouse-muscle-dataset-name) \# \<\<\<--- REPLACE with your actual HF repo URL
 
7
  multilinguality: monolingual
8
  pretty_name: Mouse Skeletal Muscle Aging Atlas (sn/scRNA-seq)
9
  size_categories:
10
+ - 10K<n<100K # Based on ~96K cells, this category fits
11
  source_datasets:
12
  - original
13
  tags:
 
43
  This dataset offers an unprecedented opportunity to:
44
  * Identify **age-specific molecular signatures** within various skeletal muscle cell types (e.g., muscle stem cells, fibroblasts, immune cells).
45
  * Uncover how cellular processes like muscle regeneration, metabolism, inflammation, and cellular senescence change with age at the single-cell level.
46
+ * Discover **biomarkers** or **therapeutic targets** related to age-associated muscle decline.
47
  * Investigate the contribution of different cell types to the overall aging process of skeletal muscle and their interplay.
48
  * Analyze shifts in cellular composition within the muscle tissue with advancing age.
49
 
50
+ This dataset thus serves as a powerful resource for understanding the intricate molecular mechanisms of aging within a vital mammalian tissue, with direct implications for longevity and healthspan research.
51
 
52
  ---
53
 
 
113
  * Discover biomarkers or therapeutic targets for sarcopenia and other age-related muscle pathologies.
114
 
115
  ### Machine Learning
116
+ * **Clustering:** Apply clustering algorithms (e.g., K-Means, Louvain) on `pca_embeddings.parquet` or `umap_embeddings.parquet` to identify distinct cell populations or sub-populations.
117
  * **Classification:** Build models to classify cell types, age groups (e.g., young vs. old mice), or other relevant phenotypes using `pca_embeddings.parquet` or `umap_embeddings.parquet` as features. `cell_metadata.parquet` provides the necessary labels.
118
  * **Regression:** Predict the biological age of a cell or donor based on gene expression or derived features.
119
  * **Dimensionality Reduction & Visualization:** Use the PCA and UMAP embeddings for generating 2D or 3D plots to visualize complex cell relationships and age-related trends.
120
  * **Feature Selection:** Identify key genes or principal components relevant to muscle aging processes.
121
 
122
+ ### Direct Download and Loading from Hugging Face Hub
123
  This dataset is hosted on the Hugging Face Hub, allowing for easy programmatic download and loading of its component files.
124
 
125
  ```python
126
  import pandas as pd
127
  from huggingface_hub import hf_hub_download
128
  import os
 
129
 
130
  # Define the Hugging Face repository ID and the local directory for downloads
131
+ HF_REPO_ID = "longevity-db/mouse-muscle-aging-atlas-snRNAseq"
132
  LOCAL_DATA_DIR = "downloaded_mouse_muscle_data"
133
 
134
  os.makedirs(LOCAL_DATA_DIR, exist_ok=True)
135
  print(f"Created local download directory: {LOCAL_DATA_DIR}")
136
 
137
+ # List of Parquet files to download (matching ONLY the files you have available)
138
+ parquet_files = [
139
  "expression.parquet",
140
  "gene_metadata.parquet",
141
  "cell_metadata.parquet",
 
143
  "pca_explained_variance.parquet",
144
  "umap_embeddings.parquet",
145
  "highly_variable_gene_metadata.parquet",
146
+ "gene_statistics.parquet"
 
 
147
  ]
148
 
149
  # Download each file
150
  downloaded_paths = {}
151
+ for file_name in parquet_files:
152
  try:
153
  path = hf_hub_download(repo_id=HF_REPO_ID, filename=file_name, local_dir=LOCAL_DATA_DIR)
154
  downloaded_paths[file_name] = path
155
  print(f"Downloaded {file_name} to: {path}")
156
  except Exception as e:
157
+ print(f"Warning: Could not download {file_name}. It might not be in the repository or its name differs. Error: {e}")
158
 
159
+ # Load core Parquet files into Pandas DataFrames
160
  df_expression = pd.read_parquet(downloaded_paths["expression.parquet"])
161
  df_pca_embeddings = pd.read_parquet(downloaded_paths["pca_embeddings.parquet"])
162
  df_umap_embeddings = pd.read_parquet(downloaded_paths["umap_embeddings.parquet"])
163
  df_cell_metadata = pd.read_parquet(downloaded_paths["cell_metadata.parquet"])
164
  df_gene_metadata = pd.read_parquet(downloaded_paths["gene_metadata.parquet"])
165
+ df_pca_explained_variance = pd.read_parquet(downloaded_paths["pca_explained_variance.parquet"])
166
+ df_hvg_metadata = pd.read_parquet(downloaded_paths["highly_variable_gene_metadata.parquet"])
167
+ df_gene_stats = pd.read_parquet(downloaded_paths["gene_statistics.parquet"])
168
+
169
 
170
  print("\n--- Data Loaded from Hugging Face Hub ---")
171
  print("Expression data shape:", df_expression.shape)
 
173
  print("UMAP embeddings shape:", df_umap_embeddings.shape)
174
  print("Cell metadata shape:", df_cell_metadata.shape)
175
  print("Gene metadata shape:", df_gene_metadata.shape)
176
+ print("PCA explained variance shape:", df_pca_explained_variance.shape)
177
+ print("HVG metadata shape:", df_hvg_metadata.shape)
178
+ print("Gene statistics shape:", df_gene_stats.shape)
179
 
180
 
181
  # Example: Prepare data for an age prediction model
182
  # IMPORTANT: You need to inspect `df_cell_metadata.columns` to find the actual age and cell type columns.
183
+ print("\nAvailable columns in cell_metadata.parquet (df_cell_metadata.columns):")
184
  print(df_cell_metadata.columns.tolist())
185
 
186
  # --- USER ACTION REQUIRED ---
 
198
  y_labels_age_prediction = df_cell_metadata[age_column_name]
199
  print(f"\nPrepared X (features) for age prediction with shape {X_features_age_prediction.shape} and y (labels) with shape {y_labels_age_prediction.shape}")
200
  else:
201
+ print(f"\nWarning: Column '{age_column_name}' not found in cell metadata for age prediction example. Please check your data.")
202
 
203
  # Example: Using cell type for a classification task
204
  if cell_type_column_name in df_cell_metadata.columns:
 
206
  y_labels_cell_type = df_cell_metadata[cell_type_column_name]
207
  print(f"Prepared X (features) for cell type classification with shape {X_features_cell_type.shape} and y (labels) with shape {y_labels_cell_type.shape}")
208
  else:
209
+ print(f"Warning: Column '{cell_type_column_name}' not found in cell metadata for cell type classification example. Please check your data.")
210
 
211
  # This data can then be split into train/test sets and used to train various ML models.
212
  ```
 
218
  **Mouse Muscle Ageing Cell Atlas Official Website:**
219
  [https://www.muscleageingcellatlas.org/mouse-pp/](https://www.muscleageingcellatlas.org/mouse-pp/)
220
 
221
+ If you use the `scanpy` library for any further analysis or preprocessing, please also cite Scanpy.
222
+
223
  ## Contributions
224
 
225
  This dataset was processed and prepared by:
 
230
 
231
  *Curated on June 15, 2025.*
232
 
233
+ **Hugging Face Repository:** [https://huggingface.co/datasets/longevity-db/mouse-muscle-aging-atlas-snRNAseq](https://huggingface.co/datasets/longevity-db/mouse-muscle-aging-atlas-snRNAseq)