A deep learning approach to automate refinement of somatic variant calling from cancer sequencing data
A partir de données portant sur 440 génomes tumoraux et 41 000 mutations somatiques, cette étude évalue, sur trois ensembles de données indépendantes, les performances d'une méthode basée sur l'apprentissage automatique pour automatiser le processus d'identification des variants somatiques
Cancer genomic analysis requires accurate identification of somatic variants in sequencing data. Manual review to refine somatic variant calls is required as a final step after automated processing. However, manual variant refinement is time-consuming, costly, poorly standardized, and non-reproducible. Here, we systematized and standardized somatic variant refinement using a machine learning approach. The final model incorporates 41,000 variants from 440 sequencing cases. This model accurately recapitulated manual refinement labels for three independent testing sets (13,579 variants) and accurately predicted somatic variants confirmed by orthogonal validation sequencing data (212,158 variants). The model improves on manual somatic refinement by reducing bias on calls otherwise subject to high inter-reviewer variability.
Nature Genetics , résumé, 2018