Norbert Deutsch1*, Gábor Erdős2 and Zsuzsanna Dosztányi1
1Eötvös Loránd University
2ELTE Protein Dynamics Research Group
norbert.deutsch [at] ttk.elte.hu
Abstract
Intrinsically disordered regions played key roles in cellular signaling and regulation, yet their contribution to human disease remained poorly understood. To address this, the present study analyzed nearly one million ClinVar missense variants, focusing on those located within intrinsically disordered regions. Pathogenic variants showed significant enrichment in short linear motifs and disordered binding regions, aligning with their central functional importance. To extend these insights beyond existing annotations, subsequent analysis utilizing the AlphaMissense tool uncovered localized “island-like” patterns of elevated pathogenicity within these functional regions. Leveraging these signals, a newly developed classifier prioritized predicted eukaryotic linear motifs (PEMs), revealing thousands of candidate functional sites linked to major disease classes, including neurological, cardiovascular, and cancer-associated genes. Specific case studies, including the genes POLK and FOXP2, demonstrated how this approach successfully connected genetic variation to molecular mechanisms. Ultimately, this framework provided a scalable strategy to interpret variants of uncertain significance and defined the functional constraints governing the disordered proteome.
Keywords: IDRs, Variants, AlphaMissense, SLiMs, Pathogenicity

