Many primary-tumor subregions exhibit low levels of molecular oxygen and restricted access to nutrients due to poor vascularization in the tissue, phenomenon known as hypoxia. Hypoxic tumors are able to regulate the expression of certain genes and signaling molecules in the microenvironment that shift it towards a more aggressive phenotype. The transcriptional landscape of the tumor favors malignant transformation of neighboring cells and their migration to distant sites. Herein, we focused on identifying key proteins that participate in the signaling crossroads between hypoxic environment and metastasis progression that remain poorly defined. To shed light on these mechanisms, we performed an integrated multi-omics analysis encompassing genomic/transcriptomic alterations of hypoxia-related genes and Buffa hypoxia scores across 17 pancarcinomas taken from the PanCancer Atlas project from The Cancer Genome Atlas consortium, protein–protein interactome network, shortest paths from hypoxia-related proteins to metastatic and angiogenic phenotypes, and drugs involved in current clinical trials to treat the metastatic disease. As results, we identified 30 hypoxia-related proteins highly involved in metastasis and angiogenesis. This set of proteins, validated with the MSK-MET Project, could represent key targets for developing therapies. The upregulation of mRNA was the most prevalent alteration in all cancer types. The highest frequencies of genomic/transcriptomic alterations and hypoxia score belonged to tumor stage 4 and positive metastatic status in all pancarcinomas. The most significantly associated signaling pathways were HIF-1, PI3K-Akt, thyroid hormone, ErbB, FoxO, mTOR, insulin, MAPK, Ras, AMPK, and VEGF. The interactome network revealed high-confidence interactions among hypoxic and metastatic proteins. The analysis of shortest paths revealed several ways to spread metastasis and angiogenesis from hypoxic proteins. Lastly, we identified 23 drugs enrolled in clinical trials focused on metastatic disease treatment. Six of them were involved in advanced-stage clinical trials: aflibercept, bevacizumab, cetuximab, erlotinib, ipatasertib, and panitumumab.

Background: It is imperative to identify drugs that allow treating symptoms of severe COVID-19. Respiratory failure is the main cause of death in severe COVID-19 patients, and the host inflammatory response at the lungs remains poorly understood.Methods: Therefore, we retrieved data from post-mortem lungs from COVID-19 patients and performed in-depth in silico analyses of single-nucleus RNA sequencing data, inflammatory protein interactome network, and shortest pathways to physiological phenotypes to reveal potential therapeutic targets and drugs in advanced-stage COVID-19 clinical trials.Results: Herein, we analyzed transcriptomics data of 719 inflammatory response genes across 19 cell types (116,313 nuclei) from lung autopsies. The functional enrichment analysis of the 233 significantly expressed genes showed that the most relevant biological annotations were inflammatory response, innate immune response, cytokine production, interferon production, macrophage activation, blood coagulation, NLRP3 inflammasome complex, and the TLR, JAK-STAT, NF-κB, TNF, oncostatin M signaling pathways. Subsequently, we identified 34 essential inflammatory proteins with both high-confidence protein interactions and shortest pathways to inflammation, cell death, glycolysis, and angiogenesis.Conclusion: We propose three small molecules (baricitinib, eritoran, and montelukast) that can be considered for treating severe COVID-19 symptoms after being thoroughly evaluated in COVID-19 clinical trials.

Expanding antiviral treatment options against SARS-CoV-2 remains crucial as the virus evolves rapidly and drug resistant strains have emerged. Broad spectrum host-directed antivirals (HDA) are promising therapeutic options, however the robust identification of relevant host factors by CRISPR/Cas9 or RNA interference screens remains challenging due to low consistency in the resulting hits. To address this issue, we employed machine learning based on experimental data from knockout screens and a drug screen. As gold standard, we assembled perturbed genes reducing virus replication or protecting the host cells. The machines based their predictions on features describing cellular localization, protein domains, annotated gene sets from Gene Ontology, gene and protein sequences, and experimental data from proteomics, phospho-proteomics, protein interaction and transcriptomic profiles of SARS-CoV-2 infected cells. The models reached a remarkable performance with a balanced accuracy of 0.82 (knockout based classifier) and 0.71 (drugs screen based classifier), suggesting patterns of intrinsic data consistency. The predicted host dependency factors were enriched in sets of genes particularly coding for development, morphogenesis, and neural related processes. Focusing on development and morphogenesis-associated gene sets, we found β-catenin to be central and selected PRI-724, a canonical β-catenin/CBP disruptor, as a potential HDA. PRI-724 limited infection with SARS-CoV-2 variants, SARS-CoV-1, MERS-CoV and IAV in different cell line models. We detected a concentration-dependent reduction in CPE development, viral RNA replication, and infectious virus production in SARS-CoV-2 and SARS-CoV-1-infected cells. Independent of virus infection, PRI-724 treatment caused cell cycle deregulation which substantiates its potential as a broad spectrum antiviral. Our proposed machine learning concept may support focusing and accelerating the discovery of host dependency factors and the design of antiviral therapies.

Expanding antiviral treatment options against SARS-CoV-2 remains crucial as the virus evolves under selection pressure which already led to the emergence of several drug resistant strains. Broad spectrum host-directed antivirals (HDA) are promising therapeutic options, however the robust identification of relevant host factors by CRISPR/Cas9 or RNA interference screens remains challenging due to low consistency in the resulting hits. To address this issue, we employed machine learning, based on experimental data from several knockout screens and a drug screen. We trained classifiers using genes essential for virus life cycle obtained from the knockout screens. The machines based their predictions on features describing cellular localization, protein domains, annotated gene sets from Gene Ontology, gene and protein sequences, and experimental data from proteomics, phospho-proteomics, protein interaction and transcriptomic profiles of SARS-CoV-2 infected cells. The models reached a remarkable performance suggesting patterns of intrinsic data consistency. The predicted HDF were enriched in sets of genes particularly encoding development, morphogenesis, and neural processes. Focusing on development and morphogenesis-associated gene sets, we found β-catenin to be central and selected PRI-724, a canonical β-catenin/CBP disruptor, as a potential HDA. PRI-724 limited infection with SARS-CoV-2 variants, SARS-CoV-1, MERS-CoV and IAV in different cell line models. We detected a concentration-dependent reduction in cytopathic effects, viral RNA replication, and infectious virus production in SARS-CoV-2 and SARS-CoV-1-infected cells. Independent of virus infection, PRI-724 treatment caused cell cycle deregulation which substantiates its potential as a broad spectrum antiviral. Our proposed machine learning concept supports focusing and accelerating the discovery of host dependency factors and identification of potential host-directed antivirals.