Exhaustive Human 9-mer Mutanome with NetMHCPan Scores

Two data files are included here: human_proteome_lookup_table.csv; This is a lookup table for the sources of the peptides in the second file. Each transcript in the human proteome is numbered to save on storage space using the column `sequence_id`. full_synthetic_mutanome.csv.gz; This contains information on the entire human mutanome, including the mutant peptide scored, the wild type peptide which is the source of the mutant peptide, and the location of the wild type peptide in the human proteome. For example, for a peptide AAAAAAAAG, the wild type peptide could be AAAAAAAAA, and the source would be 505_22_31 (meaning it is from sequence 505 in the lookup table above, from residues 22 - 31. Note that the wild type peptides and sources are nested, as the same mutant peptide can arise from different similar source peptides. The remaining columns are named as alleles, with NetMHCPan 4.1 associated scores. Again, note that due to the volume of data, we do not store scores for non-binders, only predicted binders with scores > 0.05. Hence, if for a given peptide all allele columns are empty, it has been predicted as a non-binder against all of these alleles. Where a score is present in the allele columns, it will always be greater than 0.05, and indicates predicted binding to the given MHC-I allele.