Mining conditions specific hub genes from RNA-Seq gene-expression data via biclustering and their application to drug discovery

Construction of a Pearson- and MIC-Based Co-expression Network to Identify Potential Cancer Genes

The weighted gene co-expression network analysis (WGCNA) method constructs co-expressed gene modules based on the linear similarity between paired gene expressions. Linear correlations are the main form of similarity between genes, however, nonlinear correlations still existed and had always been ignored. We proposed a modified network analysis method, WGCNA-P + M, which combines Pearson's correlation coefficient and the maximum information coefficient (MIC) as the similarity measures to assess the linear and nonlinear correlations between genes, respectively. Taking two real datasets, GSE44861 and liver hepatocellular carcinoma (TCGA-LIHC), as examples, we compared the gene modules constructed by WGCNA-P + M and WGCNA from four perspectives: the "Usefulness" score, GO enrichment analysis on genes in the gray module, prediction performance of the top hub gene, survival analysis and literature reports on different hub genes. The results showed that the modules obtained by WGCNA-P + M are more biological meaningful, the hub genes obtained from WGCNA-P + M have more potential cancer genes.

Efficient prediction of drug-drug interaction using deep learning models

A drug-drug interaction or drug synergy is extensively utilised for cancer treatment. However, prediction of drug-drug interaction is defined as an ill-posed problem, because manual testing is only implementable on small group of drugs. Predicting the drug-drug interaction score has been a popular research topic recently. Recently many machine learning models have proposed in the literature to predict the drug-drug interaction score efficiently. However, these models suffer from the over-fitting issue. Therefore, these models are not so-effective for predicting the drug-drug interaction score. In this work, an integrated convolutional mixture density recurrent neural network is proposed and implemented. The proposed model integrates convolutional neural networks, recurrent neural networks and mixture density networks. Extensive comparative analysis reveals that the proposed model significantly outperforms the competitive models.

Identification of 40S ribosomal protein S8 as a novel biomarker for alcohol‑associated hepatocellular carcinoma using weighted gene co‑expression network analysis

Alcohol‑associated hepatocellular carcinoma (HCC) is a subtype of HCC with poor prognosis. The present study aimed to identify key biomarkers for alcohol‑associated HCC. The gene data profiles and corresponding clinical traits of patients with alcohol‑associated HCC were downloaded from The Cancer Genome Atlas (TCGA) database. Firstly, good genes and good samples were identified, which were subsequently used to conduct weighted gene co‑expression network analysis (WGCNA). Hub genes in the significant modules were selected following Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, and from constructing a protein‑protein interaction (PPI) network. Real hub genes among hub genes were determined following progression, survival analysis and gene set enrichment analysis (GSEA), as well as reverse transcription‑quantitative PCR and immunohistochemical staining of non‑alcohol‑ and alcohol‑associated HCC samples. In total, 64 good samples of alcohol‑associated HCC with height score <160 were selected, from which 15,195 good genes were identified and used to conduct WGCNA; 8 gene co‑expressed modules were identified using WGCNA, while 3 modules (including pink, magenta and turquoise) were significantly associated with Child‑Pugh score, T‑stage and body weight. Following GO and KEGG analysis and construction of the PPI network, a total of 30 hub genes were identified in the aforementioned 3 gene co‑expressed modules, while 16 hub genes (including AURKB, BUB1, BUB1B, CCNB1, CCNB2, CDC20, CDCA8, CDK1, PLK1, RPS5, RPS7, RPS8, RPS14, RPS27, RPSA and TOP2A) were associated with the development of alcohol‑associated HCC, and had a significant prognosis value. Among these genes, only RPS8 was highly expressed in alcohol‑associated HCC, but not in non‑alcohol‑associated HCC, while RPS5 was not significantly associated in either alcohol‑ or non‑alcohol‑associated HCC. GSEA demonstrated that 10 pathways, including RNA polymerase and ribosome pathways were enriched in alcohol‑associated HCC samples where RPS8 was highly expressed. Taken together, the results of the present study demonstrate that RPS8 may be a novel biomarker for the diagnosis of patients with alcohol‑associated HCC.

Cucurbitacin E Inhibits Huh7 Hepatoma Carcinoma Cell Proliferation and Metastasis via Suppressing MAPKs and JAK/STAT3 Pathways

Cucurbitacin E (CuE), a highly oxygenated tetracyclic triterpene from Cucurbitaceae, has shown to exhibit potent cytotoxic and anti-proliferative properties against several human cancer cells. However, the underlying effects and mechanisms of CuE regarding hepatocellular carcinoma (HCC) have not been well understood. In the current study, unbiased RNA-sequencing (RNA-seq) and bioinformatics analysis was applied to elucidate the underlying molecular mechanism. CuE could significantly inhibit cell proliferation and migration of Huh7 cells, meanwhile CuE exhibited potent anti-angiogenic activity. RNA-seq analysis revealed that CuE negatively regulated 241 differentially expressed genes (DEGs) involved in multiple processes including cytoskeleton formation, angiogenesis and focal adhesion. Further analysis revealed that CuE effectually regulated diversified pharmacological signaling pathways such as MAPKs and JAK-STAT3. Our findings demonstrated the role of CuE in inhibiting proliferation and migration, providing an insight into the regulation of multiple signaling pathways as a new paradigm for anti-cancer treatment strategy.