Footprints of antigen processing boost MHC class II natural ligand predictions

Table 1 Summary of binding affinity (“Binders”) and eluted ligand (“Ligands”) data sets used in this work

Binders (upper table): data set reference name (“Reference”), data source (“Source”), MHC restriction (“Allele”), and the amount of sequences in the length range of 11 to 19 amino acids (“L11–19”). Ligands (lower table): data set reference name (“Reference”), data source (“Source”), MHC restriction (“Allele”), cell line species (“Cell”), amount of unique sequences present in the data set before filtering (“Unique”) and after filtering with GibbsCluster (“GC”), quantity of sequences in the 11–19mer range (“L11–19”), number of random negatives sequences added for training (“Random”). Note that the split of the Ooi et al. human data (DR15 Pm/DR51 Pm) was made using the GibbsCluster as described in the text

ISSN: 1756-994X