Deciphering the functional impact of genetic variation is required to understand phenotypic diversity and the molecular mechanisms of inherited disease and cancer. While millions of genetic variants are now mapped in genome sequencing projects, distinguishing functional variants remains a major challenge. Protein-coding variation can be interpreted using post-translational modification (PTM) sites that are core components of cellular signaling networks controlling molecular processes and pathways. ActiveDriverDB is an interactive proteo-genomics database that uses more than 260,000 experimentally detected PTM sites to predict the functional impact of genetic variation in disease, cancer and the human population. Using machine learning tools, we prioritize proteins and pathways with enriched PTM-specific amino acid substitutions that potentially rewire signaling networks via induced or disrupted short linear motifs of kinase binding. We then map these effects to site-specific protein interaction networks and drug targets. In the 2021 update, we increased the PTM datasets by nearly 50%, included glycosylation, sumoylation and succinylation as new types of PTMs, and updated the workflows to interpret inherited disease mutations. We added a recent phosphoproteomics dataset reflecting the cellular response to SARS-CoV-2 to predict the impact of human genetic variation on COVID-19 infection and disease course. Overall, we estimate that 16-21% of known amino acid substitutions affect PTM sites among pathogenic disease mutations, somatic mutations in cancer genomes and germline variants in the human population. These data underline the potential of interpreting genetic variation through the lens of PTMs and signaling networks. The open-source database is freely available at www.ActiveDriverDB.org.

Mutational signatures represent a genomic footprint of endogenous and exogenous mutational processes through tumor evolution. However, their functional impact on the proteome remains incompletely understood. We analyzed the protein-coding impact of single-base substitution (SBS) signatures in 12,341 cancer genomes from 18 cancer types. Stop-gain mutations (SGMs) (i.e., nonsense mutations) were strongly enriched in SBS signatures of tobacco smoking, APOBEC cytidine deaminases, and reactive oxygen species. These mutational processes alter specific trinucleotide contexts and thereby substitute serines and glutamic acids with stop codons. SGMs frequently affect cancer hallmark pathways and tumor suppressors such asTP53,FAT1, andAPC. Tobacco-driven SGMs in lung cancer correlate with smoking history and highlight a preventable determinant of these harmful mutations. APOBEC-driven SGMs are enriched in YTCA motifs and associate withAPOBEC3Aexpression. Our study exposes SGM expansion as a genetic mechanism by which endogenous and carcinogenic mutational processes directly contribute to protein loss of function, oncogenesis, and tumor heterogeneity.

Background. Metabolic dysfunction‐associated steatohepatitis (MASH) has become the leading cause of chronic liver disease, but there has been no approved pharmacotherapy to date. Methods. We used a network analysis approach to delineate protein‐protein interactions that contribute to the transition from steatosis to MASH, in order to identify and target this transition as a potential pharmacotherapeutic strategy. Acyl‐CoA thioesterase 1 (ACOT1) was identified as a critical node in the protein‐protein interaction (PPI) network of the transition from steatosis to MASH in patient samples. ACOT1 overexpression and silencing effects were tested in vivo on C57BL/6 mice exposed to high‐fat diet (HFD) and inoculated with an adenoviral system to modulate ACOT1 expression. Transcriptomic and untargeted lipidomic profiles were performed on the mouse livers. Results. ACOT1 expression was 3‐fold higher in MASH as compared to steatosis. In patient samples, ACOT1 was significantly correlated with the severity of MASH as reflected by the nonalcoholic fatty liver disease score. Experimental validation showed that downregulation of ACOT1 resulted in decreased lipid accumulation and prevention of MASH in vivo. Conversely, upregulation of ACOT1 via an adenoviral vector resulted in development of MASH, whereas control mice only developed steatosis. Lipidomic analysis revealed glycerophospholipids to be especially abundant in MASH accelerated by ACOT1 upregulation. Conclusion. These results suggest that ACOT1 contributes to the transition from steatosis to MASH through modulation of glycerophospholipid accumulation and its potential as a novel therapeutic target in MASH. This trial is registered with NCT02148471.

SARS‐CoV‐2 infection hijacks signaling pathways and induces protein–protein interactions between human and viral proteins. Human genetic variation may impact SARS‐CoV‐2 infection and COVID‐19 pathology; however, the genetic variation in these signaling networks remains uncharacterized. Here, we studied human missense single nucleotide variants (SNVs) altering phosphorylation sites modulated by SARS‐CoV‐2 infection, using machine learning to identify amino acid substitutions altering kinase‐bound sequence motifs. We found 2,033 infrequent phosphorylation‐associated SNVs (pSNVs) that are enriched in sequence motif alterations, potentially reflecting the evolution of signaling networks regulating host defenses. Proteins with pSNVs are involved in viral life cycle and host responses, including RNA splicing, interferon response (TRIM28), and glucose homeostasis (TBC1D4) with potential associations with COVID‐19 comorbidities. pSNVs disrupt CDK and MAPK substrate motifs and replace these with motifs of Tank Binding Kinase 1 (TBK1) involved in innate immune responses, indicating consistent rewiring of signaling networks. Several pSNVs associate with severe COVID‐19 and hospitalization (STARD13, ARFGEF2). Our analysis highlights potential genetic factors contributing to inter‐individual variation of SARS‐CoV‐2 infection and COVID‐19 and suggests leads for mechanistic and translational studies.