Background: High-throughput bio-OMIC technologies are producing high-dimension data from bio-samples at an ever increasing rate, whereas the training sample number in a traditional experiment remains small due to various difficulties. This “large p, small n” paradigm in the area of biomedical “big data” may be at least partly solved by feature selection algorithms, which select only features significantly associated with phenotypes. Feature selection is an NP-hard problem. Due to the exponentially increased time requirement for finding the globally optimal solution, all the existing feature selection algorithms employ heuristic rules to find locally optimal solutions, and their solutions achieve different performances on different datasets. Results: This work describes a feature selection algorithm based on a recently published correlation measurement, Maximal Information Coefficient (MIC). The proposed algorithm, McTwo, aims to select features associated with phenotypes, independently of each other, and achieving high classification performance of the nearest neighbor algorithm. Based on the comparative study of 17 datasets, McTwo performs about as well as or better than existing algorithms, with significantly reduced numbers of selected features. The features selected by McTwo also appear to have particular biomedical relevance to the phenotypes from the literature. Conclusion: McTwo selects a feature subset with very good classification performance, as well as a small feature number. So McTwo may represent a complementary feature selection algorithm for the high-dimensional biomedical datasets.

The recently emerged severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is the causative agent of ongoing global pandemic of COVID-19, may trigger immunosuppression in the early stage and overactive immune response in the late stage of infection; However, the underlying mechanisms are not well understood. Here we demonstrated that the SARS-CoV-2 nucleocapsid (N) protein dually regulated innate immune responses, i.e., the low-dose N protein suppressed type I interferon (IFN-I) signaling and inflammatory cytokines, whereas high-dose N protein promoted IFN-I signaling and inflammatory cytokines. Mechanistically, the SARS-CoV-2 N protein dually regulated the phosphorylation and nuclear translocation of IRF3, STAT1, and STAT2. Additionally, low-dose N protein combined with TRIM25 could suppress the ubiquitination and activation of retinoic acid-inducible gene I (RIG-I). Our findings revealed a regulatory mechanism of innate immune responses by the SARS-CoV-2 N protein, which would contribute to understanding the pathogenesis of SARS-CoV-2 and other SARS-like coronaviruses, and development of more effective strategies for controlling COVID-19.

The recent outbreak of the coronavirus disease-2019 (COVID-19) caused serious challenges to the human society in China and across the world. COVID-19 induced pneumonia in human hosts and carried a highly inter-person contagiousness. The COVID-19 patients may carry severe symptoms, and some of them may even die of major organ failures. This study utilized the machine learning algorithms to build the COVID-19 severeness detection model. Support vector machine (SVM) demonstrated a promising detection accuracy after 32 features were detected to be significantly associated with the COVID-19 severeness. These 32 features were further screened for inter-feature redundancies. The final SVM model was trained using 28 features and achieved the overall accuracy 0.8148. This work may facilitate the risk estimation of whether the COVID-19 patients would develop the severe symptoms. The 28 COVID-19 severeness associated biomarkers may also be investigated for their underlining mechanisms how they were involved in the COVID-19 infections.

Gene expression is regulated at the transcription and translation levels; thus, both transcription factors (TFs) and microRNAs (miRNA) play roles in regulation of gene expression. This study profiled differentially expressed mRNAs and miRNAs in gastric cancer tissues to construct a TF and miRNA co-regulatory network in order to identify altered genes in gastric cancer progression. A total of 70 cases gastric cancer and paired adjacent normal tissues were subjected to cDNA and miRNA microarray analyses. We obtained 887 up-regulated and 93 down-regulated genes and 41 down-regulated and 4 up-regulated miRNAs in gastric cancer tissues. Using the Transcriptional Regulatory Element Database, we obtained 105 genes that are regulated by the E2F family of genes and using Targetscan, miRanda, miRDB and miRWalk tools, we predicted potential targeting genes of these 45 miRNAs. We then built up the E2F-related TF and miRNA co-regulatory gene network and identified 9 hub-genes. Furthermore, we found that levels of E2F1, 2, 3, 4, 5, and 7 mRNAs associated with gastric cancer cell invasion capacity, and has associated with tumor differentiation. These data showed Overexpression of E2F mRNAs associated with gastric cancer progression.