洪媛媛
(2022-07-29 14:23):
#paper https://doi.org/10.1038/s41467-022-31765-8 Nat Commun 13, 4248 (2022). Accurate somatic variant detection using weakly supervised deep learning。肿瘤体细胞突变的calling一般使用统计学方法结合过滤条件来确定。这篇文章使用一种命名为“VarNet" 的深度学习方法，利用配对的肿瘤和正常DNA数据来确定体细胞突变。VarNet利用已知突变和非突变答案的肿瘤DNA和它配对正常DNA序列信息，将每个位点的base, base quality, mapping quality, strand bias 和 the reference base信息形成多维矩阵来训练模型，预测每个位置存在突变的概率。接着又在4套publicly available benchmark datasets比较VarNet和另外4种已发表方法，calling突变的Precision和recall能力，证明VarNet优于现有的4种方法。
Accurate somatic variant detection using weakly supervised deep learning
翻译
Abstract:
Identification of somatic mutations in tumor samples is commonly based on statistical methods in combination with heuristic filters. Here we develop VarNet, an end-to-end deep learning approach for identification of somatic variants from aligned tumor and matched normal DNA reads. VarNet is trained using image representations of 4.6 million high-confidence somatic variants annotated in 356 tumor whole genomes. We benchmark VarNet across a range of publicly available datasets, demonstrating performance often exceeding current state-of-the-art methods. Overall, our results demonstrate how a scalable deep learning approach could augment and potentially supplant human engineered features and heuristic filters in somatic variant calling.
翻译