Genome-wide normalized score: a novel algorithm to detect fetal trisomy 21 during non-invasive prenatal testing.

Area of Science:

• Genetics
• Bioinformatics
• Genomics

Background:

• Non-invasive prenatal testing (NIPT) for fetal trisomy 21 (T21) uses massively parallel shotgun sequencing (MPSS).
• Current algorithms are limited by low fetal DNA proportion and sequencing depth.
• Existing methods are susceptible to diluted fetal DNA and limited sequencing depth.

Purpose of the Study:

• To describe and evaluate a novel algorithm, genome-wide normalized score (GWNS), for aneuploidy detection.
• To compare GWNS performance against existing algorithms (Z-score, NCV).
• To assess algorithm efficacy under low fetal DNA conditions.

Main Methods:

• GWNS algorithm was developed, normalizing read counts by chromosome 21 fragment proportions.
• GWNS was compared to Z-score and NCV using theoretical approximations and simulations (86 cases).
• Algorithm performance was validated in 208 clinical samples and dilution experiments.

Main Results:

• GWNS required lower fetal DNA proportion and fewer MPSS reads than Z-score or NCV at fixed significance and power.
• In dilution experiments, GWNS achieved correct diagnosis at the lowest fetal DNA concentrations (3.83-4.75%).
• GWNS outperformed Z-score (4.75-5.22%) and NCV (6.47-8.58%) in low fetal fraction scenarios.

Conclusions:

• GWNS demonstrates comparable performance to Z-score and NCV for detecting fetal T21.
• GWNS shows improved performance over existing methods when fetal DNA fraction is low.
• The novel GWNS algorithm offers a potential advancement in NIPT for T21.