Benchmark of Computational Mutation Prediction

A collection of five human mutation datasets from online databases and the literature were downloaded and used in this study (Table 1). First, inherited disease-causing AASs annotated as DMs (damaging mutations) in the Human Gene Mutation Database [Stenson et al., 2009 (link)] (HGMD—November 2011; http://www.hgmd.org) and inherited putative functionally neutral AASs in the UniProt database [Apweiler et al., 2004 (link)] (UniProt—November 2011; http://www.uniprot.org/docs/humsavar) were downloaded and used to calculate the pathogenicity weights implemented in our weighted/species-specific method. Next, we obtained two human mutation datasets to assess the performance of FATHMM against the performance of other computational prediction algorithms previously reported in the literature: the VariBench database (VariBench—November 2011; http://bioinf.uta.fi/VariBench) used in a comprehensive review [Thusberg et al., 2011 (link)] of nine other computational prediction methods [Adzhubei et al., 2010 (link); Bao et al., 2005 (link); Bromberg and Rost, 2007 (link); Calabrese et al., 2009 (link); Capriotti et al., 2006 (link); Li et al., 2009 (link); Mort et al., 2010 (link); Ng and Henikoff, 2001 (link); Ramensky et al., 2002 (link); Thomas et al., 2003 (link)] and 267 AASs in four cancer-associated genes (BRCA1, MSH2, MLH1, and TP53) used in a recent review [Hicks et al., 2011 (link)] of four alternative computational prediction algorithms [Adzhubei et al., 2010 (link); Ng and Henikoff, 2001 (link); Reva et al., 2011 (link); Tavtigian et al., 2006 (link)]. Finally, we downloaded a human mutation dataset consisting of disease-associated and putative functionally neutral AASs from the SwissVar portal [Mottaz et al., 2010 (link)] (SwissVar—February 2011; http://swissvar.expasy.org) and performed an independent benchmark of FATHMM against eight other computational prediction algorithms [Adzhubei et al., 2010 (link); Calabrese et al., 2009 (link); Capriotti et al., 2006 (link); Ferrer-Costa et al., 2004 (link); Li et al., 2009 (link); Mort et al., 2010 (link); Ng and Henikoff, 2001 (link); Ramensky et al., 2002 (link); Thomas et al., 2003 (link)].