Variant effect has a null aa_mutation_start_offset

[julia326]

See the tesla code for what generates this error; basically, this variant's predicted effect has an aa_mutation_start_offset of None. Need to look into why that is.

2017-04-06 20:07:52,135 - tesla.cli:86 - ERROR - Could not find mutation start pos for variant Variant(contig='21', start=37140673, ref='G', alt='A', reference_name='GRCh38'), peptide VaccinePeptide(mutant_protein_fragment=MutantProteinFragment(variant=Variant(contig='21', start=37140673, ref='G', alt='A', reference_name='GRCh38'), gene_name=u'TTC3', amino_acids='YCGIGKVYLKKNKLVEMTLL', mutant_amino_acid_start_offset=12, mutant_amino_acid_end_offset=13, supporting_reference_transcripts=[Transcript(transcript_id=ENST00000354749, name=TTC3-001, gene_id=ENSG00000182670, gene_name=TTC3, biotype=protein_coding, location=21:37085790-37203112), Transcript(transcript_id=ENST00000399017, name=TTC3-003, gene_id=ENSG00000182670, gene_name=TTC3, biotype=protein_coding, location=21:37081596-37203112), Transcript(transcript_id=ENST00000540756, name=TTC3-024, gene_id=ENSG00000182670, gene_name=TTC3, biotype=protein_coding, location=21:37073281-37153323), Transcript(transcript_id=ENST00000355666, name=TTC3-201, gene_id=ENSG00000182670, gene_name=TTC3, biotype=protein_coding, location=21:37073271-37203103)], n_overlapping_reads=16, n_alt_reads=13, n_ref_reads=3, n_alt_reads_supporting_protein_sequence=7), epitope_predictions=[EpitopePrediction(allele='HLA-A*02:01', peptide_sequence='KLVEMTLL', length=8, ic50=335.8, percentile_rank=2.5, prediction_method_name='netMHCpan', overlaps_mutation=True, source_sequence='YCGIGKVYLKKNKLVEMTLL', offset=12, occurs_in_reference=False)], mutant_epitope_score=0.57442828263847712, wildtype_epitope_score=0, manufacturability_scores=ManufacturabilityScores(cterm_7mer_gravy_score=1.9000000000000001, max_7mer_gravy_score=1.9000000000000001, difficult_n_terminal_residue=False, c_terminal_cysteine=False, c_terminal_proline=False, cysteine_count=1, n_terminal_asparagine=False, asparagine_proline_bond_count=0))

[julia326]

Issue moved to hammerlab/isovar #81 via ZenHub