Automated inference of molecular mechanisms of disease from amino acid substitutions

• DOI: 10.1093/bioinformatics/btp528
• Publication Date: 2009-09-05T00:14:25Z
• Authors: Biao Li; Vidhya G. Krishnan; Matthew E. Mort; Fuxiao Xin; Kishore K. Kamati; David N. Cooper; Sean D. Mooney; Predrag Radivojac
• Keywords: 0301 basic medicine; 03 medical and health sciences; Amino Acid Substitution; Artificial Intelligence; Sequence Analysis, Protein; Molecular Sequence Data; Computational Biology; Humans; Proteins; Amino Acid Sequence; Databases, Protein; 3. Good health

Abstract Motivation: Advances in high-throughput genotyping and next generation sequencing have generated a vast amount of human genetic variation data. Single nucleotide substitutions within protein coding regions are particular importance owing to their potential give rise amino acid that affect structure function which may ultimately lead disease state. Over the last decade, number computational methods been developed predict whether such result an altered phenotype. Although these useful practice, accurate for intended purpose, they not well suited providing probabilistic estimates underlying mechanism. Results: We new model, MutPred, is based upon sequence, models changes structural features functional sites between wild-type mutant sequences. These changes, expressed as probabilities gain or loss function, can provide insight into specific molecular mechanism responsible MutPred also builds on established SIFT method but offers improved classification accuracy with respect mutations. Given conservative thresholds predicted disruption we propose generate reliable hypotheses basis ∼11% known inherited disease-causing note proportion functionally relevant residues sets cancer-associated somatic mutations higher than lesions Human Gene Mutation Database instead be characterized by disruptions structure. Availability: http://mutdb.org/mutpred Contact: [e-mail]; [e-mail]