AlaaALatif
commented
3 years ago - internal functionalities for A-Lab SARS-CoV-2 sequencing
- preparation of SARS-CoV-2 data files for public release (GISAID, GitHub, GCloud...etc.)
- enumerates all mutations found in the consensus sequences (FASTA)
- prepares associated metadata for upload to GISAID (REQUIRES INPUT METADATA TO BE IN SPECIFIC FORMAT -- CAN BE SKIPPED BY THE USER)
- automatically flags so-called suspicious mutations -- frame-shifting insertions/deletions (indels) and/or stop codon-introducing substitutions
- separates sample files into "whitelisted" samples (ready for upload) and "inspect" samples (may require manual inspection)
- user-specified list of non-conerning genes i.e. regions where uncommon mutations (frame-shifting indels, or stop codon-introducing substitutions) should not be flagged e.g. ORF8
- user-specified list of non-conerning mutations i.e. mutations (frame-shifting indels, or stop codon-introducing substitutions) that are expected and should not be flagged e.g ORF8:Q27* common to B.1.1.7 sequences
- option to include the raw aligned read files for archive upload to GCloud (BAM), allowing future retrospective analysis (CRITICALLY IMPORTANT FOR FUTURE RESEARCH ANALYSES)
Altered the call mechanics for
json2fasta.pyfor downloading and pre-processing GISAID's raw data. Significant improvement to memory usage for SITREP updates. Has NO impact onbjorn-releasefunctionality. Passed consistency tests using SITREP results before and after changes.