A measured score now sorts thousands of RAD51D repair changes
A cell study scored 6,888 coding changes in RAD51D, a protein used when a break cuts both strands of DNA. Harmful versions of the gene are tied to inherited breast and ovarian cancer, and most missense changes on a clinical test are still uncertain. Among missense variants, 4,593 scored neutral, 1,038 lost function and 536 fell in between. Known harmful and harmless controls separated cleanly. The score adds a functional measurement; family history and the clinical file still sit beside it.
Science··Midday
Cells under a drug pressure scored every coding change
RAD51D encodes a protein used in homologous recombination, the repair route cells use when a break cuts both strands of DNA. Pathogenic variants are linked to hereditary risk of breast cancer and ovarian cancer. A peer-reviewed study tested 6,888 coding variants for loss of function, and the same count and the same gene are the subject of the Italian report. Most missense changes found on a clinical test remain variants of uncertain significance: the DNA sequence differs, yet the evidence does not yet show whether the change is harmless or tied to disease.[1], [2]
Researchers calculated functional scores from cellular responses to olaparib. Scores were divided into neutral, intermediate and loss-of-function categories, with an additional statistical significance threshold for the intermediate and loss-of-function classes. Among measured missense variants, 4,593 were neutral, 1,038 showed loss of function and 536 were intermediate. The Italian account gives that same three-way split for the amino-acid changes.[1], [2]
Known harmful and harmless controls landed apart
A control set assembled from ClinVar contained 211 variants. All 37 pathogenic or likely pathogenic controls scored as loss of function, while all 174 benign or likely benign controls scored as neutral. That agreement describes the control set used in this experiment. Researchers also examined 70 clinically identified variants individually for homologous recombination and protein interactions. Repair proficiency correlated with the multiplexed functional scores at 0.75.[1]
The tested cells had no working copy of RAD51D. Researchers inserted thousands of different versions and exposed the cells to olaparib, a drug that inhibits PARP proteins. Cells whose homologous recombination is impaired are especially vulnerable to olaparib, so their response separated versions that left RAD51D effective from versions that compromised it. Changes in the DNA-binding region or the ATPase core reduced homologous recombination most severely. The screen also located the RAD51D-RAD51C contact inside the BCDX2 complex.[2]
The score still sits beside the rest of the file
In ClinVar the authors found 789 RAD51D missense variants of uncertain significance or with conflicting interpretations. After variants with a possible splicing effect or an intermediate functional result were set aside, 672 variants could be read more clearly. Of those, 87 percent showed normal function and 85 showed loss of function. The assay was calibrated to supply strong evidence for or against a pathogenic effect inside clinical classification.[2]
The functional score still sits beside the rest of variant classification. That classification combines how often a variant appears in families, the clinical record, molecular predictions and other independent observations. The map supplies a functional measurement for thousands of RAD51D changes. Turning that measurement into a clinical decision still requires those other lines of evidence. The screen uses a coding sequence, so some RNA splicing changes stay outside it, and the scores were read as contributing evidence rather than a personal risk figure.[1], [2]
