Machine learning‐enhanced noninvasive prenatal testing of monogenic disorders

Authors:
Publisher website
Google Scholar
Abstract
ObjectiveSingle-nucleotide variants (SNVs) are of great significance in prenatal diagnosis as they are the leading cause of inherited single-gene disorders (SGDs). Identifying SNVs in a non-invasive prenatal screening (NIPS) scenario is particularly challenging for maternally inherited SNVs. We present an improved method to predict inherited SNVs from maternal or paternal origin in a genome-wide manner.MethodsWe performed SNV-NIPS based on the combination of fragments of cell free DNA (cfDNA) features, Bayesian inference and a machine-learning (ML) prediction refinement step using random forest (RF) classifiers trained on millions of non-pathogenic variants. We next evaluate the real-world performance of our refined method in a clinical setting by testing our models on 16 families with singleton pregnancies and varying fetal fraction (FF) levels, and validate the results over millions of inherited variants in each fetus.ResultsThe average area under the ROC curve (AUC) values are 0.996 over all families for paternally inherited variants, 0.81 for the challenging maternally inherited variants, 0.86 for homozygous biallelic variants and 0.95 for compound heterozygous variants. Discriminative AUCs were achieved even in families with a low FF. We further investigate the performance of our method in correctly predicting SNVs in coding regions of clinically relevant genes and demonstrate significantly improved AUCs in these regions. Finally, we focus on the pathogenic variants in our cohort and show that our method correctly predicts if the fetus is unaffected or affected in all (10/10, 100%) of the families containing a pathogenic SNV.ConclusionsOverall, we demonstrate our ability to perform genome-wide NIPS for maternal and homozygous biallelic variants and showcase the utility of our method in a clinical setting. What is Already Known about This Topic?Current non-invasive prenatal screening (NIPS) methods have limitations in detecting maternally inherited single-nucleotide variants (SNVs) due to challenges in differentiating fetal and maternal cell-free DNA. Existing approaches for SNV detection in NIPS often focus on targeted genes or paternally inherited variants, leaving a gap in comprehensive genome-wide solutions, especially for challenging cases such as homozygous biallelic variants.What This Study Adds?This study presents an innovative method combining fragmentomic analysis, Bayesian inference, and machine learning, enabling accurate genome-wide prediction of both paternally and maternally inherited SNVs. It introduces a robust approach for detecting challenging cases, including maternally inherited and homozygous biallelic variants, in real-world clinical scenarios, enhancing the accuracy and scope of non-invasive prenatal screening.

This publication has 23 references indexed in Scilit: