Abstract:OBJECTIVES: Cell-free DNA based non-invasive prenatal testing (NIPT) has been shown with closeto-diagnosis sensitivity and specificity in detecting fetal T21, T18, and T13. Future development of this technology focuses on detection of other chromosomal aneuploidies, and more importantly chromosome copy number variants (CNVs) and single-gene mutations. Validation studies were carried out to demonstrate the feasibility of detecting CNVs and single-gene disease by exploiting the cell-free DNA testing strategy and existing NIPT platforms. METHOD: For testing CNVs, 35 plasma samples with known fetal CNVs (>10 Mb in size) were recruited, together with 1077 plasma samples with normal fetal karyotypes as the control. The samples were blindly tested using the routine NIPT testing pipelines adopted at the clinical laboratories of BGI, and analyzed with the proprietary FCAPS protocol. Results were compared to the karyotyping results to calculate the sensitivity and specificity of the NIPT test. For testing single-gene disease, we developed a unique haplotype-assisted approach which was validated in a Maple Syrup Urine Disease (MSUD) case. In a trio family (father, mother in early pregnancy, and proband), targeted sequencing (180X) of the known genes causing MSUD was used to identify the disease causative mutations in the proband child and parents. At 13 gestational weeks, maternal plasma DNA and genomic DNA of the trio family were tested with a customized array at the sequencing depth of 40X and 20X respectively to retrieve the SNPs information for the purpose of parental haplotypes deduction. Linkage analysis and Hidden Markov Model were then used to construct the fetal genotype and haplotype flanking the MSUD mutation region. RESULTS: For CNV detection, with on average 4–5 M reads per sample and the existing NIPT platforms, we were able to use the FCAPS analytic protocol to identify 32 positive samples in total of 35, and generated 6 false positive cases, leading to overall sensitivity and specificity of 91.4% and 99.4% respectively. False positive rate and false negative rate of the test were 0.56% and 0.28% respectively. We also showed that with an increased sequencing depth, the detection power of the test can be further improved. In the MUSD case, deep targeted sequencing in the trio family identified that the proband child carries a missense mutation and a large deletion on BCKDHA gene, which were inherited from the mother and father respectively. With the SNP markers of the proband and parents, parental haplotype of chromosome 19 in 6148 loci were recovered. After fetal haplotype construction, the same alleles as that of the proband in the causative gene BCKDHA were inherited to the fetus, indicating that the fetus is a MSUD patient. This result was later confirmed by invasive diagnosis, and led to a well-informed decision after post-test counselling. CONCLUSIONS: We demonstrate the successful application of NIPT in detection of CNVs larger than 10 Mb without changing much to the current pipelines of aneuploidy detection. It has the potential to be integrated into the current NIPT test and offer extra benefits to the clinical service. The haplotype-assisted approach we developed requires minimum genetic information to accurately predict fetal condition. It has the clinical potential for noninvasive prenatal diagnosis of single gene disease for family with a proband child and seeks a second healthy child.

CNValidatron, automated validation of CNV calls using computer vision

Accuracy Of Cnv Detection From Gwas Data

DeepCNV: a deep learning approach for authenticating copy number variations

CNV-P: a machine-learning framework for predicting high confident copy number variations

Copy Number Aberrations from Affymetrix SNP 6.0 Genotyping Data-How Accurate Are Commonly Used Prediction Approaches?

Evaluation of Copy Number Variation Detection for a SNP Array Platform

Noninvasive detection of chromosomal CNV and single gene disease by massively parallel sequencing of cfDNA

AutoCNV: a semiautomatic CNV interpretation system based on the 2019 ACMG/ClinGen Technical Standards for CNVs

PEcnv: accurate and efficient detection of copy number variations of various lengths

CNVpytor: a tool for copy number variation detection and analysis from read depth and allele imbalance in whole-genome sequencing

CNVoyant a machine learning framework for accurate and explainable copy number variant classification

CNV-Finder: Streamlining Copy Number Variation Discovery

ParseCNV Integrative Copy Number Variation Association Software with Quality Tracking

SeqCNV: a Novel Method for Identification of Copy Number Variations in Targeted Next-Generation Sequencing Data

A novel signal processing approach for the detection of copy number variations in the human genome

SCCNV: A Software Tool for Identifying Copy Number Variation From Single-Cell Whole-Genome Sequencing

CNVbd: A Method for Copy Number Variation Detection and Boundary Search

Robust Regression Analysis of Copy Number Variation Data based on a Univariate Score

Accurate detection of CNV based on single-nucleotide variants recalibration and image classification from whole genome sequencing

CNVABNN: An AdaBoost algorithm and neural networks-based detection of copy number variations from NGS data

Rapid, Reliable, and Interpretable CNV Curation Visualizations for Diagnostic Settings with SeeNV