Skip to main content

Table 2 Prediction of the disease type by protein similarity

From: Predicting cancer-associated germline variations in proteins

% Id % prot Sp(C) Sp (N) Sn (C) Sn (N) MCC Q2
≤30 10 0.13 0.94 0.25 0.87 0.09 0.83
≤40 75 0.26 0.85 0.29 0.83 0.11 0.74
≤50 95 0.28 0.87 0.41 0.80 0.18 0.73
≤60 99 0.34 0.88 0.47 0.82 0.26 0.76
≤70 100 0.35 0.89 0.46 0.83 0.26 0.77
≤80 100 0.36 0.89 0.48 0.83 0.29 0.78
≤90 100 0.36 0.89 0.48 0.83 0.29 0.78
  1. %prot= percentage of proteins that can be annotated with a given similarity threshold cut-off. %Id= Threshold cut-off of the sequence identity of the best hit retrieved upon a BLAST search in our dataset. For a definition of classes and scoring indexes see section: Measuring the performance.