iaode 0.2.3__tar.gz → 0.2.6__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -6,3 +6,5 @@ recursive-include iaode *.py
6
6
  recursive-exclude * __pycache__
7
7
  recursive-exclude * *.py[co]
8
8
  recursive-exclude * .DS_Store
9
+ recursive-exclude * .ipynb_checkpoints
10
+ prune */.ipynb_checkpoints
@@ -1,9 +1,9 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: iaode
3
- Version: 0.2.3
3
+ Version: 0.2.6
4
4
  Summary: Interpretable Autoencoder with Ordinary Differential Equations for single-cell omics analysis
5
5
  Author-email: Zeyu Fu <fuzeyu99@126.com>
6
- License: MIT
6
+ License-Expression: MIT
7
7
  Project-URL: Homepage, https://github.com/PeterPonyu/iAODE
8
8
  Project-URL: Documentation, https://iaode.readthedocs.io
9
9
  Project-URL: Repository, https://github.com/PeterPonyu/iAODE
@@ -49,16 +49,16 @@ Dynamic: license-file
49
49
 
50
50
  # iAODE: Interpretable Accessibility ODE VAE for scATAC-seq
51
51
 
52
- [![PyPI version](https://badge.fury.io/py/iaode.svg)](https://badge.fury.io/py/iaode) [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
52
+ [![PyPI version](https://badge.fury.io/py/iaode.svg)](https://pypi.org/project/iaode/) [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) [![PyTorch >=1.10](https://img.shields.io/badge/PyTorch-%3E=1.10-EE4C2C?logo=pytorch)](https://pytorch.org/) [![Tests](https://github.com/PeterPonyu/iAODE/actions/workflows/test.yml/badge.svg)](https://github.com/PeterPonyu/iAODE/actions/workflows/test.yml)
53
53
 
54
- `iAODE` (interpretable Accessibility ODE VAE) is a lightweight deep learning framework purpose-built for single‑cell ATAC‑seq (scATAC‑seq) data. It couples a variational autoencoder (VAE) with a Neural ODE and an interpretable bottleneck to jointly achieve:
54
+ **iAODE** (Interpretable Accessibility ODE VAE) is a lightweight deep learning framework purpose-built for single‑cell ATAC‑seq (scATAC‑seq) data. It integrates a Variational Autoencoder (VAE) with a Neural ODE and an interpretable bottleneck to jointly achieve:
55
55
 
56
- 1. **Robust modeling** of sparse count accessibility profiles (NB / ZINB likelihoods)
57
- 2. **Continuous trajectory inference** (Neural ODE pseudotime + velocity)
58
- 3. **Interpretable latent factors** (biologically aligned bottleneck)
59
- 4. **Scalable preprocessing** (TF‑IDF + highly variable peak selection) aligned with Signac / SnapATAC2 best practices
56
+ 1. **Robust Modeling:** Handles sparse count accessibility profiles using Negative Binomial (NB) or Zero-Inflated Negative Binomial (ZINB) likelihoods.
57
+ 2. **Continuous Trajectory Inference:** Infers dynamics via Neural ODE pseudotime and latent velocity fields.
58
+ 3. **Interpretable Latent Factors:** Utilizes a biologically aligned bottleneck to extract meaningful features.
59
+ 4. **Scalable Preprocessing:** Implements TF‑IDF normalization and Highly Variable Peak (HVP) selection, aligned with Signac and SnapATAC2 best practices.
60
60
 
61
- The design targets the unique characteristics of scATAC data: extreme sparsity, library size variability, heterogeneous peak accessibility kinetics, and dynamic regulatory trajectories.
61
+ The design specifically targets the unique characteristics of scATAC data, including extreme sparsity, library size variability, heterogeneous peak accessibility kinetics, and dynamic regulatory trajectories.
62
62
 
63
63
  ---
64
64
 
@@ -79,16 +79,16 @@ Encoder (MLP / Residual / Transformer) --> q(z|x)
79
79
  Decoder (NB / ZINB / MSE) -> Reconstruction x_hat
80
80
  ```
81
81
 
82
- **Key Components**:
82
+ **Key Components:**
83
83
 
84
- - **TF‑IDF Normalization**: Stabilizes cell‑wise peak depth, emphasizes specific accessibility
85
- - **Highly Variable Peaks (HVP)**: Variance / VMR / deviance‑based selection
86
- - **NB/ZINB Likelihoods**: Models over‑dispersion & zero inflation
87
- - **Neural ODE**: Smooth accessibility progression with pseudotime & velocity
88
- - **Interpretable Bottleneck**: Linear compression preserving biologically decodable axes
89
- - **Multi‑Objective Regularization**: β‑VAE KL, β‑TC, DIP, InfoVAE MMD, ODE consistency
84
+ - **TF‑IDF Normalization:** Stabilizes cell‑wise sequencing depth and highlights specific accessibility patterns.
85
+ - **Highly Variable Peaks (HVP):** Selects informative peaks using variance, VMR, or deviance‑based methods.
86
+ - **NB/ZINB Likelihoods:** Explicitly models over‑dispersion and zero inflation inherent in single-cell data.
87
+ - **Neural ODE:** Learns smooth accessibility progressions using pseudotime and velocity.
88
+ - **Interpretable Bottleneck:** A linear compression layer that preserves biologically decodable axes.
89
+ - **Multi‑Objective Regularization:** Includes $\beta$-VAE KL, $\beta$-TC, DIP, InfoVAE MMD, and ODE consistency losses.
90
90
 
91
- **Loss Function**:
91
+ **Loss Function:**
92
92
 
93
93
  ```python
94
94
  Loss = recon + i_recon + ODE_consistency + β·KL + dip·DIP + tc·TC + info·MMD
@@ -120,7 +120,7 @@ pip install -e .
120
120
  - Scanpy ≥ 1.8.0
121
121
  - scvi-tools ≥ 0.16.0
122
122
 
123
- See `requirements.txt` for complete dependencies.
123
+ See `requirements.txt` for the complete list of dependencies.
124
124
 
125
125
  ---
126
126
 
@@ -171,10 +171,10 @@ print(f"Retained {adata.n_vars} highly variable peaks")
171
171
  model = iaode.agent(
172
172
  adata,
173
173
  layer='counts',
174
- latent_dim=32, # Higher for scATAC complexity
175
- hidden_dim=512, # Deeper for regulatory patterns
174
+ latent_dim=32, # Higher dimension for scATAC complexity
175
+ hidden_dim=512, # Deeper network for regulatory patterns
176
176
  encoder_type='mlp',
177
- loss_mode='zinb', # Best for sparse scATAC data
177
+ loss_mode='zinb', # Recommended for sparse scATAC data
178
178
  use_ode=False
179
179
  )
180
180
 
@@ -265,13 +265,13 @@ plt.show()
265
265
 
266
266
  ### scATAC-seq Peak Annotation Pipeline
267
267
 
268
- **Automatic data download** - iAODE automatically downloads and caches datasets:
268
+ **Automatic Data Download** - iAODE automatically downloads and caches datasets:
269
269
 
270
270
  ```python
271
271
  import iaode
272
272
 
273
- # Downloads mouse brain 5k scATAC-seq + GENCODE vM25 GTF
274
- # Files cached in ~/.iaode/data/ and reused on subsequent runs
273
+ # Download mouse brain 5k scATAC-seq + GENCODE vM25 GTF
274
+ # Files are cached in ~/.iaode/data/ and reused on subsequent runs
275
275
  h5_file, gtf_file = iaode.datasets.mouse_brain_5k_atacseq()
276
276
 
277
277
  # Run complete annotation pipeline
@@ -279,21 +279,21 @@ adata = iaode.annotation_pipeline(
279
279
  h5_file=str(h5_file),
280
280
  gtf_file=str(gtf_file),
281
281
  promoter_upstream=2000, # TSS upstream region
282
- promoter_downstream=500, # TSS downstream region
283
- apply_tfidf=True, # TF-IDF normalization
282
+ promoter_downstream=500, # TSS downstream region
283
+ apply_tfidf=True, # Apply TF-IDF normalization
284
284
  select_hvp=True, # Select highly variable peaks
285
285
  n_top_peaks=20000 # Number of HVPs to retain
286
286
  )
287
287
 
288
- # AnnData now contains:
289
- # - adata.var['peak_type']: promoter/exonic/intronic/intergenic
288
+ # The returned AnnData object contains:
289
+ # - adata.var['peak_type']: Annotation (promoter/exonic/intronic/intergenic)
290
290
  # - adata.var['gene_name']: Associated gene names
291
291
  # - adata.var['distance_to_tss']: Distance to nearest TSS
292
292
  # - adata.var['highly_variable']: HVP selection mask
293
- # - Preprocessed with TF-IDF and HVP selection
293
+ # - Preprocessed counts (TF-IDF)
294
294
  ```
295
295
 
296
- **Available datasets:**
296
+ **Available Datasets:**
297
297
 
298
298
  ```python
299
299
  # Mouse brain 5k scATAC-seq
@@ -307,12 +307,11 @@ iaode.datasets.list_cached_files() # Show cached files
307
307
  iaode.datasets.clear_cache() # Clear all cached data
308
308
  ```
309
309
 
310
-
311
310
  ---
312
311
 
313
312
  ## Examples
314
313
 
315
- Comprehensive examples are available in the `examples/` directory. See **[examples/README.md](examples/README.md)** for detailed documentation.
314
+ Comprehensive examples are provided in the `examples/` directory. See **[examples/README.md](examples/README.md)** for detailed documentation.
316
315
 
317
316
  ### Example Scripts
318
317
 
@@ -325,7 +324,6 @@ Comprehensive examples are available in the `examples/` directory. See **[exampl
325
324
  | **`trajectory_inference_atac.py`** | Neural ODE trajectory with scATAC-seq | scATAC | Chromatin accessibility dynamics |
326
325
  | **`trajectory_inference_rna.py`** | Neural ODE trajectory with scRNA-seq | scRNA | Transcriptional trajectory |
327
326
 
328
-
329
327
  ### Running Examples
330
328
 
331
329
  ```bash
@@ -344,7 +342,7 @@ python model_evaluation_atac.py
344
342
  python trajectory_inference_rna.py
345
343
  ```
346
344
 
347
- All examples save outputs to `examples/outputs/<example_name>/` by default. See **[examples/README.md](examples/README.md)** for detailed usage instructions and customization options.
345
+ By default, all examples save outputs to `examples/outputs/<example_name>/`.
348
346
 
349
347
  ---
350
348
 
@@ -357,13 +355,13 @@ All examples save outputs to `examples/outputs/<example_name>/` by default. See
357
355
  ```python
358
356
  model = iaode.agent(
359
357
  adata, # AnnData object
360
- layer='counts', # Data layer to use
358
+ layer='counts', # Data layer to use (key in adata.layers or 'X')
361
359
  latent_dim=10, # Latent space dimension
362
360
  hidden_dim=128, # Hidden layer dimension
363
- i_dim=None, # Interpretable bottleneck dim (required if use_ode=True)
361
+ i_dim=2, # Interpretable bottleneck dimension (default: 2)
364
362
  use_ode=False, # Enable Neural ODE
365
363
  loss_mode='nb', # Loss function: 'mse', 'nb', 'zinb'
366
- encoder_type='mlp', # Encoder: 'mlp', 'residual_mlp', 'transformer', 'linear'
364
+ encoder_type='mlp', # Encoder: 'mlp', 'mlp_residual', 'transformer', 'linear'
367
365
  lr=1e-4, # Learning rate
368
366
  batch_size=128, # Batch size
369
367
  beta=1.0, # KL divergence weight
@@ -613,15 +611,15 @@ If you use iAODE in your research, please cite:
613
611
 
614
612
  ## Contributing
615
613
 
616
- Contributions are welcome! Please:
614
+ Contributions are welcome! Please follow these steps:
617
615
 
618
- 1. Fork the repository
619
- 2. Create a feature branch (`git checkout -b feature/AmazingFeature`)
620
- 3. Commit changes (`git commit -m 'Add AmazingFeature'`)
621
- 4. Push to branch (`git push origin feature/AmazingFeature`)
622
- 5. Open a Pull Request
616
+ 1. Fork the repository.
617
+ 2. Create a feature branch (`git checkout -b feature/AmazingFeature`).
618
+ 3. Commit your changes (`git commit -m 'Add AmazingFeature'`).
619
+ 4. Push to the branch (`git push origin feature/AmazingFeature`).
620
+ 5. Open a Pull Request.
623
621
 
624
- For major changes, please open an issue first to discuss proposed changes.
622
+ For major changes, please open an issue first to discuss the proposed modifications.
625
623
 
626
624
  ---
627
625
 
@@ -633,20 +631,20 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
633
631
 
634
632
  ## Acknowledgments
635
633
 
636
- Built upon ideas from:
634
+ Built upon concepts from:
637
635
 
638
- - **scVI-tools**: scVI, PEAKVI, POISSONVI architectures
639
- - **Signac** and **SnapATAC2**: scATAC-seq best practices
640
- - **Neural ODE** literature: Continuous latent dynamics
636
+ - **scVI-tools:** scVI, PEAKVI, and POISSONVI architectures.
637
+ - **Signac** and **SnapATAC2:** Best practices for scATAC-seq data processing.
638
+ - **Neural ODE:** Literature regarding continuous latent dynamics.
641
639
 
642
- PyTorch, AnnData, and Scanpy ecosystems provide the foundation.
640
+ Foundational support is provided by the PyTorch, AnnData, and Scanpy ecosystems.
643
641
 
644
642
  ---
645
643
 
646
644
  ## Contact
647
645
 
648
- - **GitHub Issues**: [https://github.com/PeterPonyu/iAODE/issues](https://github.com/PeterPonyu/iAODE/issues)
649
- - **Email**: <fuzeyu99@126.com>
646
+ - **GitHub Issues:** [https://github.com/PeterPonyu/iAODE/issues](https://github.com/PeterPonyu/iAODE/issues)
647
+ - **Email:** <fuzeyu99@126.com>
650
648
 
651
649
  ---
652
650
 
@@ -1,15 +1,15 @@
1
1
  # iAODE: Interpretable Accessibility ODE VAE for scATAC-seq
2
2
 
3
- [![PyPI version](https://badge.fury.io/py/iaode.svg)](https://badge.fury.io/py/iaode) [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
3
+ [![PyPI version](https://badge.fury.io/py/iaode.svg)](https://pypi.org/project/iaode/) [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) [![PyTorch >=1.10](https://img.shields.io/badge/PyTorch-%3E=1.10-EE4C2C?logo=pytorch)](https://pytorch.org/) [![Tests](https://github.com/PeterPonyu/iAODE/actions/workflows/test.yml/badge.svg)](https://github.com/PeterPonyu/iAODE/actions/workflows/test.yml)
4
4
 
5
- `iAODE` (interpretable Accessibility ODE VAE) is a lightweight deep learning framework purpose-built for single‑cell ATAC‑seq (scATAC‑seq) data. It couples a variational autoencoder (VAE) with a Neural ODE and an interpretable bottleneck to jointly achieve:
5
+ **iAODE** (Interpretable Accessibility ODE VAE) is a lightweight deep learning framework purpose-built for single‑cell ATAC‑seq (scATAC‑seq) data. It integrates a Variational Autoencoder (VAE) with a Neural ODE and an interpretable bottleneck to jointly achieve:
6
6
 
7
- 1. **Robust modeling** of sparse count accessibility profiles (NB / ZINB likelihoods)
8
- 2. **Continuous trajectory inference** (Neural ODE pseudotime + velocity)
9
- 3. **Interpretable latent factors** (biologically aligned bottleneck)
10
- 4. **Scalable preprocessing** (TF‑IDF + highly variable peak selection) aligned with Signac / SnapATAC2 best practices
7
+ 1. **Robust Modeling:** Handles sparse count accessibility profiles using Negative Binomial (NB) or Zero-Inflated Negative Binomial (ZINB) likelihoods.
8
+ 2. **Continuous Trajectory Inference:** Infers dynamics via Neural ODE pseudotime and latent velocity fields.
9
+ 3. **Interpretable Latent Factors:** Utilizes a biologically aligned bottleneck to extract meaningful features.
10
+ 4. **Scalable Preprocessing:** Implements TF‑IDF normalization and Highly Variable Peak (HVP) selection, aligned with Signac and SnapATAC2 best practices.
11
11
 
12
- The design targets the unique characteristics of scATAC data: extreme sparsity, library size variability, heterogeneous peak accessibility kinetics, and dynamic regulatory trajectories.
12
+ The design specifically targets the unique characteristics of scATAC data, including extreme sparsity, library size variability, heterogeneous peak accessibility kinetics, and dynamic regulatory trajectories.
13
13
 
14
14
  ---
15
15
 
@@ -30,16 +30,16 @@ Encoder (MLP / Residual / Transformer) --> q(z|x)
30
30
  Decoder (NB / ZINB / MSE) -> Reconstruction x_hat
31
31
  ```
32
32
 
33
- **Key Components**:
33
+ **Key Components:**
34
34
 
35
- - **TF‑IDF Normalization**: Stabilizes cell‑wise peak depth, emphasizes specific accessibility
36
- - **Highly Variable Peaks (HVP)**: Variance / VMR / deviance‑based selection
37
- - **NB/ZINB Likelihoods**: Models over‑dispersion & zero inflation
38
- - **Neural ODE**: Smooth accessibility progression with pseudotime & velocity
39
- - **Interpretable Bottleneck**: Linear compression preserving biologically decodable axes
40
- - **Multi‑Objective Regularization**: β‑VAE KL, β‑TC, DIP, InfoVAE MMD, ODE consistency
35
+ - **TF‑IDF Normalization:** Stabilizes cell‑wise sequencing depth and highlights specific accessibility patterns.
36
+ - **Highly Variable Peaks (HVP):** Selects informative peaks using variance, VMR, or deviance‑based methods.
37
+ - **NB/ZINB Likelihoods:** Explicitly models over‑dispersion and zero inflation inherent in single-cell data.
38
+ - **Neural ODE:** Learns smooth accessibility progressions using pseudotime and velocity.
39
+ - **Interpretable Bottleneck:** A linear compression layer that preserves biologically decodable axes.
40
+ - **Multi‑Objective Regularization:** Includes $\beta$-VAE KL, $\beta$-TC, DIP, InfoVAE MMD, and ODE consistency losses.
41
41
 
42
- **Loss Function**:
42
+ **Loss Function:**
43
43
 
44
44
  ```python
45
45
  Loss = recon + i_recon + ODE_consistency + β·KL + dip·DIP + tc·TC + info·MMD
@@ -71,7 +71,7 @@ pip install -e .
71
71
  - Scanpy ≥ 1.8.0
72
72
  - scvi-tools ≥ 0.16.0
73
73
 
74
- See `requirements.txt` for complete dependencies.
74
+ See `requirements.txt` for the complete list of dependencies.
75
75
 
76
76
  ---
77
77
 
@@ -122,10 +122,10 @@ print(f"Retained {adata.n_vars} highly variable peaks")
122
122
  model = iaode.agent(
123
123
  adata,
124
124
  layer='counts',
125
- latent_dim=32, # Higher for scATAC complexity
126
- hidden_dim=512, # Deeper for regulatory patterns
125
+ latent_dim=32, # Higher dimension for scATAC complexity
126
+ hidden_dim=512, # Deeper network for regulatory patterns
127
127
  encoder_type='mlp',
128
- loss_mode='zinb', # Best for sparse scATAC data
128
+ loss_mode='zinb', # Recommended for sparse scATAC data
129
129
  use_ode=False
130
130
  )
131
131
 
@@ -216,13 +216,13 @@ plt.show()
216
216
 
217
217
  ### scATAC-seq Peak Annotation Pipeline
218
218
 
219
- **Automatic data download** - iAODE automatically downloads and caches datasets:
219
+ **Automatic Data Download** - iAODE automatically downloads and caches datasets:
220
220
 
221
221
  ```python
222
222
  import iaode
223
223
 
224
- # Downloads mouse brain 5k scATAC-seq + GENCODE vM25 GTF
225
- # Files cached in ~/.iaode/data/ and reused on subsequent runs
224
+ # Download mouse brain 5k scATAC-seq + GENCODE vM25 GTF
225
+ # Files are cached in ~/.iaode/data/ and reused on subsequent runs
226
226
  h5_file, gtf_file = iaode.datasets.mouse_brain_5k_atacseq()
227
227
 
228
228
  # Run complete annotation pipeline
@@ -230,21 +230,21 @@ adata = iaode.annotation_pipeline(
230
230
  h5_file=str(h5_file),
231
231
  gtf_file=str(gtf_file),
232
232
  promoter_upstream=2000, # TSS upstream region
233
- promoter_downstream=500, # TSS downstream region
234
- apply_tfidf=True, # TF-IDF normalization
233
+ promoter_downstream=500, # TSS downstream region
234
+ apply_tfidf=True, # Apply TF-IDF normalization
235
235
  select_hvp=True, # Select highly variable peaks
236
236
  n_top_peaks=20000 # Number of HVPs to retain
237
237
  )
238
238
 
239
- # AnnData now contains:
240
- # - adata.var['peak_type']: promoter/exonic/intronic/intergenic
239
+ # The returned AnnData object contains:
240
+ # - adata.var['peak_type']: Annotation (promoter/exonic/intronic/intergenic)
241
241
  # - adata.var['gene_name']: Associated gene names
242
242
  # - adata.var['distance_to_tss']: Distance to nearest TSS
243
243
  # - adata.var['highly_variable']: HVP selection mask
244
- # - Preprocessed with TF-IDF and HVP selection
244
+ # - Preprocessed counts (TF-IDF)
245
245
  ```
246
246
 
247
- **Available datasets:**
247
+ **Available Datasets:**
248
248
 
249
249
  ```python
250
250
  # Mouse brain 5k scATAC-seq
@@ -258,12 +258,11 @@ iaode.datasets.list_cached_files() # Show cached files
258
258
  iaode.datasets.clear_cache() # Clear all cached data
259
259
  ```
260
260
 
261
-
262
261
  ---
263
262
 
264
263
  ## Examples
265
264
 
266
- Comprehensive examples are available in the `examples/` directory. See **[examples/README.md](examples/README.md)** for detailed documentation.
265
+ Comprehensive examples are provided in the `examples/` directory. See **[examples/README.md](examples/README.md)** for detailed documentation.
267
266
 
268
267
  ### Example Scripts
269
268
 
@@ -276,7 +275,6 @@ Comprehensive examples are available in the `examples/` directory. See **[exampl
276
275
  | **`trajectory_inference_atac.py`** | Neural ODE trajectory with scATAC-seq | scATAC | Chromatin accessibility dynamics |
277
276
  | **`trajectory_inference_rna.py`** | Neural ODE trajectory with scRNA-seq | scRNA | Transcriptional trajectory |
278
277
 
279
-
280
278
  ### Running Examples
281
279
 
282
280
  ```bash
@@ -295,7 +293,7 @@ python model_evaluation_atac.py
295
293
  python trajectory_inference_rna.py
296
294
  ```
297
295
 
298
- All examples save outputs to `examples/outputs/<example_name>/` by default. See **[examples/README.md](examples/README.md)** for detailed usage instructions and customization options.
296
+ By default, all examples save outputs to `examples/outputs/<example_name>/`.
299
297
 
300
298
  ---
301
299
 
@@ -308,13 +306,13 @@ All examples save outputs to `examples/outputs/<example_name>/` by default. See
308
306
  ```python
309
307
  model = iaode.agent(
310
308
  adata, # AnnData object
311
- layer='counts', # Data layer to use
309
+ layer='counts', # Data layer to use (key in adata.layers or 'X')
312
310
  latent_dim=10, # Latent space dimension
313
311
  hidden_dim=128, # Hidden layer dimension
314
- i_dim=None, # Interpretable bottleneck dim (required if use_ode=True)
312
+ i_dim=2, # Interpretable bottleneck dimension (default: 2)
315
313
  use_ode=False, # Enable Neural ODE
316
314
  loss_mode='nb', # Loss function: 'mse', 'nb', 'zinb'
317
- encoder_type='mlp', # Encoder: 'mlp', 'residual_mlp', 'transformer', 'linear'
315
+ encoder_type='mlp', # Encoder: 'mlp', 'mlp_residual', 'transformer', 'linear'
318
316
  lr=1e-4, # Learning rate
319
317
  batch_size=128, # Batch size
320
318
  beta=1.0, # KL divergence weight
@@ -564,15 +562,15 @@ If you use iAODE in your research, please cite:
564
562
 
565
563
  ## Contributing
566
564
 
567
- Contributions are welcome! Please:
565
+ Contributions are welcome! Please follow these steps:
568
566
 
569
- 1. Fork the repository
570
- 2. Create a feature branch (`git checkout -b feature/AmazingFeature`)
571
- 3. Commit changes (`git commit -m 'Add AmazingFeature'`)
572
- 4. Push to branch (`git push origin feature/AmazingFeature`)
573
- 5. Open a Pull Request
567
+ 1. Fork the repository.
568
+ 2. Create a feature branch (`git checkout -b feature/AmazingFeature`).
569
+ 3. Commit your changes (`git commit -m 'Add AmazingFeature'`).
570
+ 4. Push to the branch (`git push origin feature/AmazingFeature`).
571
+ 5. Open a Pull Request.
574
572
 
575
- For major changes, please open an issue first to discuss proposed changes.
573
+ For major changes, please open an issue first to discuss the proposed modifications.
576
574
 
577
575
  ---
578
576
 
@@ -584,20 +582,20 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
584
582
 
585
583
  ## Acknowledgments
586
584
 
587
- Built upon ideas from:
585
+ Built upon concepts from:
588
586
 
589
- - **scVI-tools**: scVI, PEAKVI, POISSONVI architectures
590
- - **Signac** and **SnapATAC2**: scATAC-seq best practices
591
- - **Neural ODE** literature: Continuous latent dynamics
587
+ - **scVI-tools:** scVI, PEAKVI, and POISSONVI architectures.
588
+ - **Signac** and **SnapATAC2:** Best practices for scATAC-seq data processing.
589
+ - **Neural ODE:** Literature regarding continuous latent dynamics.
592
590
 
593
- PyTorch, AnnData, and Scanpy ecosystems provide the foundation.
591
+ Foundational support is provided by the PyTorch, AnnData, and Scanpy ecosystems.
594
592
 
595
593
  ---
596
594
 
597
595
  ## Contact
598
596
 
599
- - **GitHub Issues**: [https://github.com/PeterPonyu/iAODE/issues](https://github.com/PeterPonyu/iAODE/issues)
600
- - **Email**: <fuzeyu99@126.com>
597
+ - **GitHub Issues:** [https://github.com/PeterPonyu/iAODE/issues](https://github.com/PeterPonyu/iAODE/issues)
598
+ - **Email:** <fuzeyu99@126.com>
601
599
 
602
600
  ---
603
601
 
@@ -1,10 +1,6 @@
1
- import gc
2
- import os
3
1
  import time
4
2
  import numpy as np
5
- import pandas as pd # type: ignore
6
3
  import scanpy as sc # type: ignore
7
- import scipy.sparse as sp # type: ignore
8
4
  import torch
9
5
  from sklearn.cluster import KMeans # type: ignore
10
6
  from sklearn.metrics import ( # type: ignore
@@ -77,7 +73,7 @@ class DataSplitter:
77
73
  self.test_idx = test_idx
78
74
  self.train_val_idx = train_val_idx
79
75
 
80
- print(f"\nData split sizes:")
76
+ print("\nData split sizes:")
81
77
  print(f" Total: {self.n_samples}")
82
78
  print(f" Train: {len(train_idx)} ({len(train_idx)/self.n_samples*100:.1f}%)")
83
79
  print(f" Val: {len(val_idx)} ({len(val_idx)/self.n_samples*100:.1f}%)")
@@ -245,7 +241,7 @@ def train_scvi_models(adata, splitter, n_latent=10, n_epochs=400, batch_size=128
245
241
  'actual_epochs': actual_epochs
246
242
  }
247
243
 
248
- print(f"✓ SCVI training completed")
244
+ print("✓ SCVI training completed")
249
245
  print(f" Epochs: {actual_epochs}/{n_epochs}, Time: {train_time:.2f}s, Peak GPU Memory: {peak_memory:.3f} GB")
250
246
 
251
247
  except Exception as e:
@@ -308,7 +304,7 @@ def train_scvi_models(adata, splitter, n_latent=10, n_epochs=400, batch_size=128
308
304
  'actual_epochs': actual_epochs
309
305
  }
310
306
 
311
- print(f"✓ PEAKVI training completed")
307
+ print("✓ PEAKVI training completed")
312
308
  print(f" Epochs: {actual_epochs}/{n_epochs}, Time: {train_time:.2f}s, Peak GPU Memory: {peak_memory:.3f} GB")
313
309
 
314
310
  except Exception as e:
@@ -371,7 +367,7 @@ def train_scvi_models(adata, splitter, n_latent=10, n_epochs=400, batch_size=128
371
367
  'actual_epochs': actual_epochs
372
368
  }
373
369
 
374
- print(f"✓ POISSONVI training completed")
370
+ print("✓ POISSONVI training completed")
375
371
  print(f" Epochs: {actual_epochs}/{n_epochs}, Time: {train_time:.2f}s, Peak GPU Memory: {peak_memory:.3f} GB")
376
372
 
377
373
  except Exception as e:
@@ -1,7 +1,6 @@
1
1
  import numpy as np
2
2
  import pandas as pd # type: ignore
3
3
  from sklearn.decomposition import PCA # type: ignore
4
- from scipy.stats import entropy # type: ignore
5
4
  from scipy.linalg import svd # type: ignore
6
5
  import warnings
7
6
 
@@ -37,7 +37,7 @@ Usage Example:
37
37
  >>> latent = model.get_latent()
38
38
  """
39
39
 
40
- __version__ = "0.2.3"
40
+ __version__ = "0.2.4"
41
41
  __author__ = "Zeyu Fu"
42
42
 
43
43
  # Import main classes and functions
@@ -10,8 +10,7 @@ import urllib.request
10
10
  import gzip
11
11
  import shutil
12
12
  from pathlib import Path
13
- from typing import Optional, Tuple
14
- import anndata as ad # type: ignore
13
+ from typing import Tuple
15
14
 
16
15
 
17
16
  def get_data_dir() -> Path:
@@ -63,7 +62,7 @@ def _download_file(url: str, output_path: Path, desc: str = "file") -> None:
63
62
 
64
63
  urllib.request.urlretrieve(url, output_path, reporthook=reporthook)
65
64
  print() # New line after progress
66
- print(f" ✓ Downloaded successfully")
65
+ print(" ✓ Downloaded successfully")
67
66
  except Exception as e:
68
67
  print(f"\n ✗ Download failed: {e}")
69
68
  raise
@@ -145,10 +144,10 @@ def mouse_brain_5k_atacseq(force_download: bool = False, use_local: bool = True)
145
144
  print(f"✓ Using local files from: {local_examples_dir}")
146
145
  # Copy to cache for consistency
147
146
  if not h5_file.exists():
148
- print(f" Copying H5 to cache...")
147
+ print(" Copying H5 to cache...")
149
148
  shutil.copy(local_h5, h5_file)
150
149
  if not gtf_file.exists():
151
- print(f" Copying GTF to cache...")
150
+ print(" Copying GTF to cache...")
152
151
  shutil.copy(local_gtf, gtf_file)
153
152
  print()
154
153
  print("=" * 70)
@@ -172,7 +171,7 @@ def mouse_brain_5k_atacseq(force_download: bool = False, use_local: bool = True)
172
171
  print("MANUAL DOWNLOAD REQUIRED")
173
172
  print("=" * 70)
174
173
  print("Please download the file manually:")
175
- print(f"\n1. Visit: https://www.10xgenomics.com/datasets/fresh-cortex-from-adult-mouse-brain-p-50-1-standard-2-0-0")
174
+ print("\n1. Visit: https://www.10xgenomics.com/datasets/fresh-cortex-from-adult-mouse-brain-p-50-1-standard-2-0-0")
176
175
  print(" Or try direct link:")
177
176
  print(f" {h5_url}")
178
177
  print(f"\n2. Save as: {h5_file}")
@@ -194,7 +193,7 @@ def mouse_brain_5k_atacseq(force_download: bool = False, use_local: bool = True)
194
193
  # Clean up compressed file to save space
195
194
  if gtf_gz.exists():
196
195
  gtf_gz.unlink()
197
- print(f" ✓ Removed compressed file to save space")
196
+ print(" ✓ Removed compressed file to save space")
198
197
  else:
199
198
  print(f"✓ Using cached GTF: {gtf_file}")
200
199
 
@@ -265,7 +264,7 @@ def human_pbmc_5k_atacseq(force_download: bool = False) -> Tuple[Path, Path]:
265
264
  # Clean up compressed file
266
265
  if gtf_gz.exists():
267
266
  gtf_gz.unlink()
268
- print(f" ✓ Removed compressed file to save space")
267
+ print(" ✓ Removed compressed file to save space")
269
268
  else:
270
269
  print(f"✓ Using cached GTF: {gtf_file}")
271
270