Better prediction of functional effects for sequence variants

• DOI: 10.1186/1471-2164-16-s8-s1
• Publication Date: 2015-06-18T13:59:33Z
• Authors: Maximilian Hecht; Yana Bromberg; Burkhard Rost
• Keywords: Sequence (biology)

Elucidating the effects of naturally occurring genetic variation is one major challenges for personalized health and medicine. Here, we introduce SNAP2, a novel neural network based classifier that improves over state-of-the-art in distinguishing between effect neutral variants. Our method's improved performance results from screening many potentially relevant protein features refining our development data sets. Cross-validated on >100k experimentally annotated variants, SNAP2 significantly outperformed other methods, attaining two-state accuracy (effect/neutral) 83%. also combinations methods. Performance increased human variants but much more so organisms. carefully calibrated reliability index informs selection experimental follow up, with most strongly predicted half all at 96% accuracy. As expected, evolutionary information automatically generated multiple sequence alignments gave strongest signal prediction. However, optimized new method to perform surprisingly well even without alignments. This feature reduces prediction runtime by two orders magnitude, enables cross-genome comparisons, renders as best solution 10-20% orphans. available at: https://rostlab.org/services/snap2web Delta, input computing difference scores native amino acid variant acid; nsSNP, non-synoymous SNP; PMD, Protein Mutant Database; SNAP, Screening non-acceptable polymorphisms; SNP, single nucleotide polymorphism; variant, any changing variant.