thekaplanlab / convart_pipeline Goto Github PK

Possible bug in the create_alignments.py.

Searching for exact sequence in convart_gene table is slow since the DB has to perform exhaustive search to find the exact match. Given the fact that protein sequence length can be arbitrarily long, it is better to store a hash in the DB and query based on that indexed column. This way, protein sequence lookups will be much much faster.

The option to select different colour schemes in the web UI will be good to have and users can try and visualise instantly rather than changing in the code.

Update: Transferred correct repository under (thekaplanlab/msabrowser#5)