Search

Many machine learning applications in bioinformatics currently rely on matching gene identities when analyzing input gene signatures and fail to take advantage of preexisting knowledge about gene functions. To further enable comparative analysis of OMICS datasets, including target deconvolution and mechanism of action studies, we develop an approach that represents gene signatures projected onto their biological functions, instead of their identities, similar to how the word2vec technique works in natural language processing. We develop the Functional Representation of Gene Signatures (FRoGS) approach by training a deep learning model and demonstrate that its application to the Broad Institutes L1000 datasets results in more effective compound-target predictions than models based on gene identities alone. By integrating additional pharmacological activity data sources, FRoGS significantly increases the number of high-quality compound-target predictions relative to existing approaches, many of which are supported by in silico and/or experimental evidence. These results underscore the general utility of FRoGS in machine learning-based bioinformatics applications. Prediction networks pre-equipped with the knowledge of gene functions may help uncover new relationships among gene signatures acquired by large-scale OMICs studies on compounds, cell types, disease models, and patient cohorts.

Abstract Aims This study compared the prevalences of metabolic syndrome and of cardiac or kidney comorbidities among patients with hepatocellular carcinoma (HCC) associated with metabolic dysfunction-related fatty liver disease (MAFLD), chronic infection with hepatitis B or C virus (HBV or HCV), or the combination of MAFLD and chronic HBV infection. Methods Medical records were retrospectively analyzed for patients with HCC who underwent hepatectomy between March 2013 and March 2023. Patients with HCC of different etiologies were compared in terms of their clinicodemographic characteristics and laboratory data before surgery. Results Of the 2422 patients, 1,822 (75.2%) were chronically infected with HBV without MAFLD and HCV, 415 (17.2%) had concurrent MAFLD and chronic HBV infection but no HCV infection, 121 (5.0%) had MAFLD without hepatitis virus infection, and 64 (2.6%) were chronically infected with HCV in the presence or absence of MAFLD and HBV infection. Compared to patients chronically infected with HBV without MAFLD and HCV, those with MAFLD but no hepatitis virus infection showed significantly lower prevalence of cirrhosis, ascites, portal hypertension, alpha-fetoprotein concentration ≥ 400 ng/mL, tumor size > 5 cm, multinodular tumors and microvascular invasion. Conversely, they showed significantly higher prevalence of metabolic syndrome, hypertension, type 2 diabetes, abdominal obesity, history of cardiovascular disease, T-wave alterations, hypertriglyceridemia and hyperuricemia, as well as higher risk of arteriosclerotic cardiovascular disease. Compared to patients with MAFLD but no hepatitis virus infection, those with concurrent MAFLD and chronic infection with HBV showed significantly higher prevalence of cirrhosis, ascites and portal hypertension, but significantly lower prevalence of hypertension and history of cardiovascular disease. Compared to patients with other etiologies, those chronically infected with HCV in the presence or absence of MAFLD and HBV infection, showed significantly higher prevalence of cirrhosis, portal hypertension, ascites, and esophagogastric varices. Conclusion Patients with HCC associated with MAFLD tend to have a background of less severe liver disease than those with HCC of other etiologies, but they may be more likely to suffer metabolic syndrome or comorbidities affecting the heart or kidneys.

ObjectiveThis retrospective cohort study aimed to investigate the composition and diversity of lung microbiota in patients with severe pneumonia and explore its association with short-term prognosis.MethodsA total of 301 patients diagnosed with severe pneumonia underwent bronchoalveolar lavage fluid metagenomic next-generation sequencing (mNGS) testing from February 2022 to January 2024. After applying exclusion criteria, 236 patients were included in the study. Baseline demographic and clinical characteristics were compared between survival and non-survival groups. Microbial composition and diversity were analyzed using alpha and beta diversity metrics. Additionally, LEfSe analysis and machine learning methods were employed to identify key pathogenic microorganism associated with short-term mortality. Microbial interaction modes were assessed through network co-occurrence analysis.ResultsThe overall 28-day mortality rate was 37.7% in severe pneumonia. Non-survival patients had a higher prevalence of hypertension and exhibited higher APACHE II and SOFA scores, higher procalcitonin (PCT), and shorter hospitalization duration. Microbial α and β diversity analysis showed no significant differences between the two groups. However, distinct species diversity patterns were observed, with the non-survival group showing a higher abundance of Acinetobacter baumannii, Klebsiella pneumoniae, and Enterococcus faecium, while the survival group had a higher prevalence of Corynebacterium striatum and Enterobacter. LEfSe analysis identified 29 distinct terms, with 10 potential markers in the non-survival group, including Pseudomonas sp. and Enterococcus durans. Machine learning models selected 16 key pathogenic bacteria, such as Klebsiella pneumoniae, significantly contributing to predicting short-term mortality. Network co-occurrence analysis revealed greater complexity in the non-survival group compared to the survival group, with differences in central genera.ConclusionOur study highlights the potential significance of lung microbiota composition in predicting short-term prognosis in severe pneumonia patients. Differences in microbial diversity and composition, along with distinct microbial interaction modes, may contribute to variations in short-term outcomes. Further research is warranted to elucidate the clinical implications and underlying mechanisms of these findings.

Abstract Background Peutz–Jeghers syndrome (PJS) is a rare genetic disorder characterized by the development of pigmented spots, gastrointestinal polyps and increased susceptibility to cancers. Currently, most studies have investigated intestinal microbiota through fecal microbiota, and there are few reports about mucosa-associated microbiota. It remains valuable to search for the key intestinal microbiota or abnormal metabolic pathways linked to PJS. Aim This study aimed to assess the structure and composition of mucosa-associated microbiota in patients with PJS and to explore the potential influence of intestinal microbiota disorders and metabolite changes on PJS. Methods The bacterial composition was analyzed in 13 PJS patients and 12 controls using 16S rRNA gene sequencing (Illumina MiSeq) for bacteria. Differential analyses of the intestinal microbiota were performed from the phylum to species level. Liquid chromatography-tandem mass spectrometry (LC‒MS) was used to detect the differentially abundant metabolites of PJS patients and controls to identify different metabolites and metabolic biomarkers of small intestinal mucosa samples. Results High-throughput sequencing confirmed the special characteristics and biodiversity of the mucosa microflora in patients with PJS. They had lower bacterial biodiversity than controls. The abundance of intestinal mucosal microflora was significantly lower than that of fecal microflora. In addition, lipid metabolism, amino acid metabolism, carbohydrate metabolism, nucleotide metabolism and other pathways were significantly different from those of controls, which were associated with the development of the enteric nervous system, intestinal inflammation and development of tumors. Conclusion This is the first report on the mucosa-associated microbiota and metabolite profile of subjects with PJS, which may be meaningful to provide a structural basis for further research on intestinal microecology in PJS.

Copper (Cu) is a trace element, essential for fish growth. In the current study, in addition to growth performance, we first explored the effects of Cu on collagen synthesis and myofiber growth and development in juvenile grass carp (Ctenopharyngodon idella). A total of 1080 fish (11.16 ± 0.01 g) were randomly divided into 6 treatments (3 replicates per treatment) to receive five doses of organic Cu, which were Cu citrate (CuCit) at 0.99 (basal diet), 2.19, 4.06, 6.15, and 8.07 mg/kg, and one dose of inorganic Cu (CuSO4·5H2O at 3.15 mg/kg), for 9 weeks. The results showed appropriate Cu level (4.06 mg/kg) enhanced growth performance, improved nutritional Cu status, and downregulated Cu-transporting ATPase 1 mRNA levels in the hepatopancreas, intestine, and muscle of juvenile grass carp. Meanwhile, collagen content in fish muscle was increased after Cu intake, which was probably due to the following pathways: (1) activating CTGF/TGF-β1/Smads signaling pathway to regulate collagen transcription; (2) upregulating of La ribonucleoprotein domain family 6 (LARP6) mRNA levels to regulate translation initiation; (3) increasing proline hydroxylase, lysine hydroxylase, and lysine oxidase activities to regulate posttranslational modifications. In addition, optimal Cu group increased myofiber diameters and the frequency of myofibers with diameter >50 μm, which might be associated with upregulation of cyclin B, cyclin D, cyclin E, proliferating cell nuclear antigen, myogenic determining factor (MyoD), myogenic factor 5, myogenin (MyoG), myogenic regulatory factor 4 and myosin heavy chain (MyHC) and downregulation of myostatin mRNA levels, increasing protein levels of MyoD, MyoG and MyHC in fish muscle. Finally, based on percentage weight gain (PWG), serum ceruloplasmin (Cp) activity and collagen content in fish muscle, Cu requirements were determined as 4.74, 4.37 and 4.62 mg/kg diet (CuCit as Cu source) of juvenile grass carp, respectively. Based on PWG and Cp activity, compared to CuSO4·5H2O, the efficacy of CuCit were 131.80% and 115.38%, respectively. Our findings provide new insights into Cu supplementation to promote muscle growth in fish, and help improve the overall productivity of aquaculture.