The GENDULF algorithm: mining transcriptomics to uncover modifier genes for monogenic diseases

• DOI: 10.15252/msb.20209701
• Publication Date: 2020-12-08T12:31:39Z
• Authors: Noam Auslander; Daniel M Ramos; Ivette Zelaya; Hiren Karathia; Thomas O. Crawford; Alejandro A Schäffer; Charlotte J Sumner; Eytan Ruppin
• Keywords: 0301 basic medicine; Medicine (General); QH301-705.5; Genetic Linkage; digenic inheritance; cystic fibrosis; 03 medical and health sciences; R5-920; Methods; Data Mining; Humans; Disease; Biology (General); Genetic Association Studies; spinal muscular atrophy; modifier gene; 0303 health sciences; Genes, Modifier; Reproducibility of Results; 3. Good health; HEK293 Cells; gene expression; Transcriptome; Algorithms

Modifier genes are believed to account for the clinical variability observed in many Mendelian disorders, but their identification remains challenging due limited availability of genomics data from large patient cohorts. Here, we present GENDULF (GENetic moDULators identiFication), one first methods facilitate prediction disease modifiers using healthy and diseased tissue gene expression data. is designed monogenic diseases which mechanism loss function leading reduced mutated gene. When applied cystic fibrosis, successfully identifies multiple, previously established modifiers, including EHF, SLC6A14, CLCA1. It then utilized spinal muscular atrophy (SMA) predicts U2AF1 as a modifier whose low correlates with higher SMN2 pre-mRNA exon 7 retention. Indeed, knockdown SMA patient-derived cells leads increased full-length transcript SMN protein expression. Taking advantage increasing transcriptomic data, novel addition existing strategies genetic providing insights into pathogenesis uncovering therapeutic targets.