Some changes in preparation for computing a minimal epitope report:

julia326 commented 7 years ago

making EpitopePrediction a class, moving logistic_epitope_score into its methods, adding a wildtype peptide to its members (reference epitope computation logic borrowed from TESLA code, see https://github.com/hammerlab/tesla/blob/df7bfdb6241899d9b4fa2bb1dc359c2ad4935979/tesla/cli.py#L81)
adding a predicted_effect() method to MutantProteinFragment, changing a couple places in the code to use that for computing the top priority effect of a variant given a mutant protein fragment; also adding a global_start_pos() method (see same TESLA code for reference)
adding a hook into the CLI to make a min epitope report, not implemented yet
setting --min-epitope-score to 0.0 in the test script so I can see multiple variants

Start of work for #142

coveralls commented 7 years ago

Coverage increased (+0.3%) to 91.096% when pulling fa5a510ee8cf680d103e583582598f86ae7c6ece on minimal-epitope-report into 30940ec48ed6e67b9c6d91dadab7cab89cd31426 on master.

iskandr commented 7 years ago

Failure on Travis:

************* Module vaxrank.mutant_protein_fragment
E:182,12: Undefined variable 'logger' (undefined-variable)

coveralls commented 7 years ago

Coverage increased (+0.7%) to 91.312% when pulling e01f4131a1fcd4478590b482c724dfd4a9c40662 on minimal-epitope-report into c7c1c8ac853d82d9cc83eb6d9974ef75f0bb0560 on master.

iskandr commented 7 years ago

LGTM

openvax / vaxrank

Some changes in preparation for computing a minimal epitope report: #141