PT - JOURNAL ARTICLE AU - Rajarshi Ghosh AU - Ninad Oak AU - Sharon E. Plon TI - Evaluation of <em>in silico</em> algorithms for use with ACMG/AMP clinical variant interpretation guidelines AID - 10.1101/146100 DP - 2017 Jan 01 TA - bioRxiv PG - 146100 4099 - http://biorxiv.org/content/early/2017/06/05/146100.short 4100 - http://biorxiv.org/content/early/2017/06/05/146100.full AB - The ACMG/AMP variant classification guidelines for clinical reporting recommend complete concordance of predictions among all in silico algorithms used without specifying the number or types of algorithms. The subjective nature of this recommendation contributes to discordance of variant classification among clinical laboratories. Using 14,819 benign or pathogenic missense variants from the ClinVar database, we compared performance of 25 algorithms across datasets differing in distinct biological and technical variables. There was wide variability in concordance among different combinations of algorithms with particularly low concordance for benign variants. We identified recently developed algorithms with high predictive power and robust to variables like disease mechanism, gene constraint and mode of inheritance, although poorer performing algorithms are more frequently used based on review of the clinical genetics literature (2011-2017). We describe high performing algorithm combinations with increased concordance in variant assertion, which should lead to more informed in silico algorithm usage by diagnostic laboratories.