A 19-Gene Signature of Serous Ovarian Cancer Identified by Machine Learning and Systems Biology: Prospects for Diagnostics and Personalized Medicine.

Ovarian cancer is a major cause of cancer deaths among women. Early diagnosis and precision/personalized medicine are essential to reduce mortality and morbidity of ovarian cancer, as with new molecular targets to accelerate drug discovery. We report here an integrated systems biology and machine learning (ML) approach based on the differential coexpression analysis to identify candidate systems biomarkers (i.e., gene modules) for serous ovarian cancer. Accordingly, four independent transcriptome datasets were statistically analyzed independently and common differentially expressed genes (DEGs) were identified. Using these DEGs, coexpressed gene pairs were unraveled. Subsequently, differential coexpression networks between the coexpressed gene pairs were reconstructed so as to identify the differentially coexpressed gene modules. Based on the established criteria, "SOV-module" was identified as being significant, consisting of 19 genes. Using independent datasets, the diagnostic capacity of the SOV-module was evaluated using principal component analysis (PCA) and ML techniques. PCA showed a sensitivity and specificity of 96.7% and 100%, respectively, and ML analysis showed an accuracy of up to 100% in distinguishing phenotypes in the present study sample. The prognostic capacity of the SOV-module was evaluated using survival and ML analyses. We found that the SOV-module's performance for prognostics was significant (p-value = 1.36 × 10-4) with an accuracy of 63% in discriminating between survival and death using ML techniques. In summary, the reported genomic systems biomarker candidate offers promise for personalized medicine in diagnosis and prognosis of serous ovarian cancer and warrants further experimental and translational clinical studies.

Drug repurposing for rheumatoid arthritis: Identification of new drug candidates via bioinformatics and text mining analysis.

Rheumatoid arthritis (RA) is an autoimmune disease that results in the destruction of tissue by attacks on the patient by his or her own immune system. Current treatment strategies are not sufficient to overcome RA. In the present study, various transcriptomic data from synovial fluids, synovial fluid-derived macrophages, and blood samples from patients with RA were analysed using bioinformatics approaches to identify tissue-specific repurposing drug candidates for RA. Differentially expressed genes (DEGs) were identified by integrating datasets for each tissue and comparing diseased to healthy samples. Tissue-specific protein-protein interaction (PPI) networks were generated and topologically prominent proteins were selected. Transcription-regulating biomolecules for each tissue type were determined from protein-DNA interaction data. Common DEGs and reporter biomolecules were used to identify drug candidates for repurposing using the hypergeometric test. As a result of bioinformatic analyses, 19 drugs were identified as repurposing candidates for RA, and text mining analyses supported our findings. We hypothesize that the FDA-approved drugs momelotinib, ibrutinib, and sodium butyrate may be promising candidates for RA. In addition, CHEMBL306380, Compound 19a (CHEMBL3116050), ME-344, XL-019, TG100801, JNJ-26483327, and NV-128 were identified as novel repurposing candidates for the treatment of RA. Preclinical and further validation of these drugs may provide new treatment options for RA.

Forecasting Gastric Cancer Diagnosis, Prognosis, and Drug Repurposing with Novel Gene Expression Signatures.

Gastric cancer (GC) is a prevalent disease worldwide with high mortality and poor treatment success. Early diagnosis of GC and forecasting of its prognosis with the use of biomarkers are directly relevant to achieve both personalized/precision medicine and innovation in cancer therapeutics. Gene expression signatures offer one of the promising avenues of research in this regard, as well as guiding drug repurposing analyses in cancers. Using publicly accessible gene expression datasets from the Gene Expression Omnibus and The Cancer Genome Atlas (TCGA), we report here original findings on co-expressed gene modules that are differentially expressed between 133 GC samples and 46 normal tissues, and thus hold potential for novel diagnostic candidates for GC. Furthermore, we found two co-expressed gene modules were significantly associated with poor survival outcomes revealed by survival analysis of the RNA-Seq TCGA datasets. We identified STAT6 (signal transducer and activator of transcription 6) as a key regulator of the identified gene modules. Finally, potential therapeutic drugs that may target and reverse the expression of the identified altered gene modules examined for drug repurposing analyses and the unraveled compounds were further investigated in the literature by the text mining method. Accordingly, we found several repurposed drug candidates, including Trichostatin A, Vorinostat, Parthenolide, Panobinostat, Brefeldin A, Belinostat, and Danusertib. Through text mining analysis and literature search validation, Belinostat and Danusertib were suggested as possible novel drug candidates for GC treatment. These findings collectively inform multiple aspects of GC medical management, including its precision diagnosis, forecasting of possible outcomes, and drug repurposing for innovation in GC medicines in the future.

Machine Learning Prediction of Resistance to Subinhibitory Antimicrobial Concentrations from Escherichia coli Genomes.

Escherichia coli is an important cause of bacterial infections worldwide, with multidrug-resistant strains incurring substantial costs on human lives. Besides therapeutic concentrations of antimicrobials in health care settings, the presence of subinhibitory antimicrobial residues in the environment and in clinics selects for antimicrobial resistance (AMR), but the underlying genetic repertoire is less well understood. Here, we used machine learning to predict the population doubling time and cell growth yield of 1,407 genetically diverse E. coli strains expanding under exposure to three subinhibitory concentrations of six classes of antimicrobials from single-nucleotide genetic variants, accessory gene variation, and the presence of known AMR genes. We predicted cell growth yields in the held-out test data with an average correlation (Spearman's ρ) of 0.63 (0.36 to 0.81 across concentrations) and cell doubling times with an average correlation of 0.59 (0.32 to 0.92 across concentrations), with moderate increases in sample size unlikely to improve predictions further. This finding points to the remaining missing heritability of growth under antimicrobial exposure being explained by effects that are too rare or weak to be captured unless sample size is dramatically increased, or by effects other than those conferred by the presence of individual single-nucleotide polymorphisms (SNPs) and genes. Predictions based on whole-genome information were generally superior to those based only on known AMR genes and were accurate for AMR resistance at therapeutic concentrations. We pinpointed genes and SNPs determining the predicted growth and thereby recapitulated many known AMR determinants. Finally, we estimated the effect sizes of resistance genes across the entire collection of strains, disclosing the growth effects for known resistance genes in each individual strain. Our results underscore the potential of predictive modeling of growth patterns from genomic data under subinhibitory concentrations of antimicrobials, although the remaining missing heritability poses a challenge for achieving the accuracy and precision required for clinical use. IMPORTANCE Predicting bacterial growth from genome sequences is important for a rapid characterization of strains in clinical diagnostics and to disclose candidate novel targets for anti-infective drugs. Previous studies have dissected the relationship between bacterial growth and genotype in mutant libraries for laboratory strains, yet no study so far has examined the predictive power of genome sequence in natural strains. In this study, we used a high-throughput phenotypic assay to measure the growth of a systematic collection of natural Escherichia coli strains and then employed machine learning models to predict bacterial growth from genomic data under nontherapeutic subinhibitory concentrations of antimicrobials that are common in nonclinical settings. We found a moderate to strong correlation between predicted and actual values for the different collected data sets. Moreover, we observed that the known resistance genes are still effective at sublethal concentrations, pointing to clinical implications of these concentrations.

Identification of Common Pathogenetic Processes between Schizophrenia and Diabetes Mellitus by Systems Biology Analysis.

Schizophrenia (SCZ) is a psychiatric disorder characterized by both positive symptoms (i.e., psychosis) and negative symptoms (such as apathy, anhedonia, and poverty of speech). Epidemiological data show a high likelihood of early onset of type 2 diabetes mellitus (T2DM) in SCZ patients. However, the molecular processes that could explain the epidemiological association between SCZ and T2DM have not yet been characterized. Therefore, in the present study, we aimed to identify underlying common molecular pathogenetic processes and pathways between SCZ and T2DM. To this aim, we analyzed peripheral blood mononuclear cell (PBMC) transcriptomic data from SCZ and T2DM patients, and we detected 28 differentially expressed genes (DEGs) commonly modulated between SCZ and T2DM. Inflammatory-associated processes and membrane trafficking pathways as common biological processes were found to be in common between SCZ and T2DM. Analysis of the putative transcription factors involved in the regulation of the DEGs revealed that STAT1 (Signal Transducer and Activator of Transcription 1), RELA (v-rel reticuloendotheliosis viral oncogene homolog A (avian)), NFKB1 (Nuclear Factor Kappa B Subunit 1), and ERG (ETS-related gene) are involved in the expression of common DEGs in SCZ and T2DM. In conclusion, we provide core molecular signatures and pathways that are shared between SCZ and T2DM, which may contribute to the epidemiological association between them.