HLA gene algorithm predicts recurrent miscarriage risk
Researchers including a team from the Institute of Human Immunology in Warsaw built an algorithm called IMMATCH that analyzed HLA gene haplotypes from couples with histories of successful pregnancies or recurrent miscarriage. Using high-resolution typing and linear algebra on peptide binding affinity data, the algorithm retrospectively predicted recurrent miscarriage with an AUC of 0.71 (p = 0.0035). With an adjustable threshold, it could reach 86% sensitivity. The work, published via doi.org, sets out a method for calculating genetic risk of recurrent miscarriage from paternal and maternal HLA data.
The retrospective model could inform future tools for unexplained infertility, but it has not been validated prospectively or clinically.
Source: Towards a gamete matching platform: using immunogenetics and artificial intelligence to predict recurrent miscarriage. (doi.org).
Written by the Genomes desk from the primary source linked above and checked against it. Research use only; not medical advice. Corrections: [email protected].
- UK Biobank expands genetic analysis to diverse ancestry groups September 18, 2025
- FinnGen study finds 30 new disease associations in Finnish population January 18, 2023
- GWAS Catalog adds sequencing data, links to Polygenic Score Catalog January 1, 2023
- Common gene variants change how a rare cleft palate mutation plays out July 1, 2022
- Polygenic scores lose accuracy fast across UK Biobank ancestry groups January 1, 2022