Multiple z-Score Based Method for Noninvasive Prenatal Test Using Cell-Free DNA in Maternal Plasma

Abstract

Objective: To improve the detecting accuracy of chromosomal aneuploidy of fetus by non-invasive prenatal testing (NIPT) using next generation sequencing data of pregnant women's cell-free DNA. Methods: We proposed the multi-Z method which uses 21 z-scores for each autosomal chromosome to detect aneuploidy of the chromosome, while the conventional NIPT method uses only one z-score. To do this, mapped read numbers of a certain chromosome were normalized by those of the other 21 chromosomes. Average and standard deviation (SD), which are used for calculating z-score of each sample, were obtained with normalized values between all autosomal chromosomes of control samples. In this way, multiple z-scores can be calculated for 21 autosomal chromosomes except oneself. Results: Multi-Z method showed 100% sensitivity and specificity for 187 samples sequenced to 3 M reads while the conventional NIPT method showed 95.1% specificity. Similarly, for 216 samples sequenced to 1 M reads, Multi-Z method showed 100% sensitivity and 95.6% specificity and the conventional NIPT method showed a result of 75.1% specificity. Conclusion: Multi-Z method showed higher accuracy and robust results than the conventional method even at low coverage reads.

Improving the accuracy of noninvasive prenatal testing through size‐selection between fetal and maternal cfDNA

“…We adopted the multi z-score method used in previous studies to detect fetal aneuploidies for specific chromosomes. 31 We created a heatmap plot to visualize the multi z-scores (Figures 3-5). The We also confirmed the quality of the sequencing data by assessing the distribution of cfDNA fragments.…”

“…We adopted the multi z‐score method used in previous studies to detect fetal aneuploidies for specific chromosomes 31 . We created a heatmap plot to visualize the multi z‐scores (Figures 3–5).…”

Noninvasive Prenatal Testing for Fetal Chromosomal Abnormalities Using Massively Parallel Sequencing: Clinical Experience from 7910 Korean Pregnancies

“…To overcome the above limitation in conventional methods, we developed two algorithms: Double-Z score algorithm and Multi-Z algorithm [8]. Double-Z algorithm calculates a z-score within a target sample, by comparing the number of GC corrected reads across all chromosomes.…”