Deep convolutional neural networks for accurate somatic mutation detection.

Authors: Sahraeian, Sayed Mohammad Ebrahim and Liu, Ruolin and Lau, Bayo and Podesta, Karl and Mohiyuddin, Marghoob and Lam, Hugo Y K

Accurate detection of somatic mutations is still a challenge in cancer analysis. Here we present NeuSomatic, the first convolutional neural network approach for somatic mutation detection, which significantly outperforms previous methods on different sequencing platforms, sequencing strategies, and tumor purities. NeuSomatic summarizes sequence alignments into small matrices and incorporates more than a hundred features to capture mutation signals effectively. It can be used universally as a stand-alone somatic mutation detection method or with an ensemble of existing methods to achieve the highest accuracy.

Journal: Nature communications
DOI: 10.1038/s41467-019-09027-x
Year: 2019

Read publication