Prediction of pathological mutations in proteins: the challenge of integrating sequence conservation and structure stability principles

摘要: The recent drop in genome sequencing costs has created a promising horizon for the development of genomic medicine. Within biomedical environment, data are increasingly used disease diagnosis and prognosis, treatment development, counseling, so on. Many these applications rely on identification causing variants. This is particularly challenging problem because large number wide variety sequence variants identified projects, also we only have limited understanding physicochemical/biochemical properties that differentiate neutral from pathologic Nonetheless, last years witnessed important methodologicaladvancesforoneclassofvariants,thosecorrespondingtochanges amino-acid proteins. Proteins main constituent living systems. We know although their biological essentially determined by sequence, not all changes this same impact. Some neutral, but others affect protein function lead to disease. A body evidence shows whether one or other case depends such as mutation location structure, interspecies conservation, Mutation prediction methods based features good success rates, 70‐90% range, representation over time suggests there performance plateau would limit applicability. In light most advances field, after reviewing foundations methods, discuss existence threshold how it can be overcomed. C � 2013 John Wiley & Sons, Ltd.