nominally significantp <0.05
They first extracted nominally significant signals ( p <0.05) from three GWAS studies (WTCCC, German, and NIMH GAIN datasets; Fangerau et al., 2004 ; Wellcome Trust Case Control Consortium, 2007 ; Baum et al., 2008 ), and prioritized potential candidates based on independent, converging lines of evidence from various bioinformatics resources.