Roche Sequencing Solutions, Belmont, CA, 94002, USA.
Microsoft Azure, Dublin 18, D18 P521, Ireland.
Nat Commun. 2019 Mar 4;10(1):1041. doi: 10.1038/s41467-019-09027-x.
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.
准确检测体细胞突变在癌症分析中仍然是一个挑战。在这里，我们提出了 NeuSomatic，这是第一个用于体细胞突变检测的卷积神经网络方法，它在不同的测序平台、测序策略和肿瘤纯度上都显著优于以前的方法。NeuSomatic 将序列比对总结成小矩阵，并结合了一百多种特征来有效地捕获突变信号。它可以作为一种独立的体细胞突变检测方法普遍使用，也可以与现有的方法集合使用，以达到最高的准确性。
