用于对种系变异致病性进行分类的深度学习模型的表型评估。
Phenotypic evaluation of deep learning models for classifying germline variant pathogenicity.
发表日期：2024 Oct 19
作者：
Ryan D Chow, Katherine L Nathanson, Ravi B Parikh
来源：
npj Precision Oncology
摘要：
用于预测变异致病性的深度学习模型尚未针对现实世界的临床表型进行彻底评估。在这里，我们将最先进的致病性预测模型应用于英国生物银行参与者的遗传性乳腺癌基因变异。模型预测 BRCA1、BRCA2 和 PALB2（而非 ATM 和 CHEK2）的错义变异与乳腺癌风险相关。然而，当专门应用于意义不确定的变异时，深度学习模型的临床实用性有限。© 2024。作者。
Deep learning models for predicting variant pathogenicity have not been thoroughly evaluated on real-world clinical phenotypes. Here, we apply state-of-the-art pathogenicity prediction models to hereditary breast cancer gene variants in UK Biobank participants. Model predictions for missense variants in BRCA1, BRCA2 and PALB2, but not ATM and CHEK2, were associated with breast cancer risk. However, deep learning models had limited clinical utility when specifically applied to variants of uncertain significance.© 2024. The Author(s).