Automated prediction of the clinical impact of structural copy number variations

Abstract

Copy number variants (CNVs) play an important role in many biological processes, including the development of genetic diseases, making them attractive targets for genetic analyses. The interpretation of the effect of these structural variants is a challenging problem due to highly variable numbers of gene, regulatory, or other genomic elements affected by the CNV. This led to the demand for the interpretation tools that would relieve researchers, laboratory diagnosticians, genetic counselors, and clinical geneticists from the laborious process of annotation and classification of CNVs. We designed and validated a prediction method (ISV; Interpretation of Structural Variants) that is based on boosted trees which takes into account annotations of CNVs from several publicly available databases. The presented approach achieved more than 98% prediction accuracy on both copy number loss and copy number gain variants while also allowing CNVs being assigned “uncertain” significance in predictions. We believe that ISV’s prediction capability and explainability have a great potential to guide users to more precise interpretations and classifications of CNVs.

Combination of Expert Guidelines-based and Machine Learning-based Approaches Leads to Superior Accuracy of Automated Prediction of Clinical Effect of Copy Number Variations

“…In the process of interpretation of small DNA sequence variants 1 several in silico tools are widely used. Similar fully automated tools, such as ISV [6][7][8] , that use prediction methods based on machine learning algorithms to evaluate the clinical impact of CNVs have also been developed.…”

“…ISV (Interpretation of Structural Variation) is a machine learning-based approach in which the pathogenicity of a candidate CNV is assessed by observing the counts of overlapped subcategories of genes and regulatory elements 8 . ISV is based on boosted trees and it classifies most variants reliably and with a high accuracy (~98%) among the compared tools 8 .…”

“…These professional standards tend to encourage consistency and transparency of CNV evaluation across clinical laboratories. In another approach, CNV classification is based on the fully automated machine learning algorithms that can go beyond the established scheme, thus having a great potential to further improve the prediction accuracy [6][7][8] . The CNVs' pathogenicity effect can be also evaluated according to aggregation of per-base single nucleotide polymorphism (SNP) pathogenicity scores within the CNV intervals 9 .…”

“…Another available method for pathogenicity prediction is X-CNV 7 , the model trained using the XGBoost classifier. Recently, we have developed a tool called ISV (Interpretation of Structural Variants) 8 which predicts the probability of pathogenicity of a CNV and its likely clinical impact based on specified CNV coordinates (see the ISV subsection in Methods for details).…”

“…Several in silico CNV pathogenicity prediction tools have been reported recently differing in the specific machine learning algorithms they implement, in the usage of specific genomic elements as features for model training, and in the parameters applied for model training. One of the tools is the ISV (Interpretation of Structural Variants) 8 which predicts the probability of pathogenicity of a CNV and its likely clinical impact based on specified CNV coordinates (see the ISV subsection in Methods for details). One of the tools is StrVCTVRE 6 focusing on SVs ’ overlapping exons.…”