Systematic prediction of degrons and E3 ubiquitin ligase binding via deep learning

• DOI: 10.1186/s12915-022-01364-6
• Publication Date: 2022-07-14T10:08:06Z
• Authors: Chao Hou; Yuxuan Li; Mengyao Wang; Hong Wu; Tingting Li
• Keywords: Degron; Proteome; Protein Degradation; Sequence motif

Abstract Background Degrons are short linear motifs, bound by E3 ubiquitin ligase to target protein substrates be degraded the ubiquitin-proteasome system. Mutations leading deregulation of degron functionality disrupt control abundance due mistargeting proteins destined for degradation and often result in pathologies. Targeting degrons small molecules also emerges as an exciting drug design strategy upregulate expression specific proteins. Despite their essential function disease targetability, reliable identification remains a conundrum. Here, we developed deep learning-based model named Degpred that predicts general directly from sequences. Results We showed BERT-based performed well predicting singly Then, used learning predict proteome-widely. successfully captured typical degron-related sequence properties predicted beyond those motif-based methods which use handful motifs match possible degrons. Furthermore, calculated using on our collected E3-substrate interaction dataset constructed regulatory network assigning E3s with motifs. Critically, experimentally verified SPOP binding CBX6 prompts mediates SPOP. system is important tumorigenesis surveying mutations TCGA. Conclusions provides efficient tool proteome-wide prediction captures previously methods, thus greatly expanding landscape, should advance understanding degradation, allow exploration uncharacterized alterations diseases. To make it easier readers access datasets, integrated these data into website http://degron.phasep.pro/ .