- Updated the fetch_refseq_sequence function to handle HTTP errors more robustly by distinguishing between urllib and requests exceptions.
- Added input validation to skip empty or NaN RefSeq IDs.
- Cleaned up whitespace and comma-separated values from the RefSeq ID.
- Enhanced error messages for better debugging.
- Updated exception handling in _fetch_sequence_for_row to accommodate new error handling in fetch_refseq_sequence.
Update README.md: Clarify application description and improve example input/output formatting
Enhance my_utils.py: Add docstrings for randomSVM and randomSearch functions to improve code documentation
Update notebook files: Modify versioning information for 01_EDA_Psort.ipynb and 04_Training.ipynb
Remove outdated environment configuration files and update README with detailed usage instructions for the Protein Location Predictor application. Add error handling for model loading in prediction utility, and include a sample protein sequence in FASTA format. Update the project structure and installation steps to reflect the latest changes.
- Updated embedding loading function to return processed embeddings and accession identifiers.
- Enhanced confusion matrix visualization with improved aesthetics.
- Consolidated PCA, t-SNE, and UMAP plotting functions for better clarity and consistency.
- Added LabelEncoder for encoding target labels in Random Forest and SVM training functions.
- Increased timeout for sequence fetching functions to improve reliability.
- Removed unused imports and cleaned up code for better readability.
- Updated documentation for functions to clarify parameters and return types.
Update project structure and remove unused notebooks
- Added `*/embeddings` to .gitignore to exclude embedding directories.
- Updated `EDA_Psort.ipynb` with new file size for version control.
- Deleted obsolete notebooks: `ESMC_300m.ipynb`, `ESMC_600m.ipynb`, and `ProstT5.ipynb`.
- Introduced `Get_embeddings.ipynb` to streamline embedding processes.
- Enhanced `my_utils.py` with error handling for embedding functions.
Add Tkinter-based GUI for Protein Location Prediction
- Implemented a GUI in GUI.py for loading FASTA files and running protein location prediction tools.
- Added functionality to load FASTA files and display messages for user feedback.
- Integrated prediction functions for PROST and ESM models (300m and 600m).
- Created a menu with options to load files, run predictions, and exit the application.
- Included buttons for each prediction method and organized layout for user interaction.
