v1.1.0

1.1.0 - All Changes

Added

• Dependency on Biopython, specifically the SeqIO module.
• Example of a FASTA file containing multiple sequences, with example/multi_seqs.fas.
• Added line to README.md regarding the RCSB PDB no longer providing updated secondary structure flat files (i.e. ss.txt.gz).
• Added descriptions of new output options.
• example/1qys_ss_conf.fas included as an example of adding confidence scores to the .fas format.
• example/multi_seqs.fas for examples of predicting from a file containing more than one sequence.

Changed

• Behavior of utilities.py and run_model.py so they now parse and predict multiple sequences from a single FASTA file.
• utilities.py uses a new wrapper around the Biopython parser.
• run_model.py refactored to predict from multiple sequences in a single FASTA files.

To Do

• Shift to using the Biopython FASTA file parser.
• Update run_model.py to run and save batches of sequences from a single input FASTA file.
• Add name and ID post-processing after Biopython read to improve the downstream file naming.
• Add separate inference script for large dataset processing with batches.