Predicting regional somatic mutation rates using DNA motifs.

How the locus-specificity of epigenetic modifications is regulated remains an unanswered question. A contributing mechanism is that epigenetic enzymes are recruited to specific loci by DNA binding factors recognizing particular sequence motifs (referred to as epi-motifs). Using these motifs to predict biological outputs depending on local epigenetic state such as somatic mutation rates would confirm their functionality. Here, we used DNA motifs including known TF motifs and epi-motifs as a surrogate of epigenetic signals to predict somatic mutation rates in 13 cancers at an average 23kbp resolution. We implemented an interpretable neural network model, called contextual regression, to successfully learn the universal relationship between mutations and DNA motifs, and uncovered motifs that are most impactful on the regional mutation rates such as TP53 and epi-motifs associated with H3K9me3. Furthermore, we identified genomic regions with significantly higher mutation rates than the expected values in each individual tumor and demonstrated that such cancer-related regions can accurately predict cancer types. Interestingly, we found that the same mutation signatures often have different contributions to cancer-related and cancer-independent regions, and we also identified the motifs with the most contribution to each mutation signature.

Transcriptome and lipidome profile of human mesenchymal stem cells with reduced senescence and increased trilineage differentiation ability upon drug treatment.

Human Mesenchymal stem cells (hMSCs) are multi-potential cells which are widely used in cell therapy. However, the frequently emerged senescence and decrease of differentiation capabilities limited the broad applications of MSC. Several strategies such as small molecules treatment have been widely studied and used to improve the stem characteristics bypassing the senescence but the exact mechanisms for them to reduce senescence have not been fully studied. In this study, hMSCs were treated by rapamycin, oltipraz, metformin, and vitamin C for the indicated time and these cells were subjected to senescence evaluation and trilineage differentiation. Furthermore, transcriptomics and lipidomics datasets of hMSCs after drug treatment were analyzed to interpret biological pathways responsible for their anti-senescence effects. Although four drugs exhibited significant activities in promoting MSC osteogenic differentiation, metformin is the optimal drug to promote trilineage differentiation. GO terms illustrated that the anti-aging effects of drugs were mainly associated with cellular senescence, mitotic and meiosis process. Biosynthesis of phosphatidylcholines (PC) and phosphatidylethanolamine (PE) were inhibited whereas production of phosphatidylinositols (PIs) and saturated fatty acids (SFA)/ mono-unsaturated fatty acids (MUFA) conversion was activated. Medium free fatty acids (FFA) was increased in hMSCs with different anti-aging phenotypes. Therefore, we established a comprehensive method in assessing drug intervention based on the results of transcriptomics and lipidomics. The method can be used to study different biological phenotypes upon drug intervention in MSC which will extend the clinical application of hMSCs.

A Novel Transcriptome Integrated Network Approach Identifies the Key Driver lncRNA Involved in Cell Cycle With Chromium (VI)-Treated BEAS-2B Cells.

Hexavalent chromium [Cr(VI)] is a well-known occupational carcinogen, but the mechanisms contributing to DNA damage and cell cycle alternation have not been fully characterized. To study the dose-response effects of Cr(VI) on transcription, we exposed BEAS-2B cells to Cr(VI) at concentrations of 0.2, 0.6, and 1.8 µmol/L for 24 h. Here, we identified 1,484 differentially expressed genes (DEGs) in our transcript profiling data, with the majority of differentially expressed transcripts being downregulated. Our results also showed that these DEGs were enriched in pathways associated with the cell cycle, including DNA replication, chromatin assembly, and DNA repair. Using the differential expressed genes related to cell cycle, a weighted gene co-expression network was constructed and a key mRNA-lncRNA regulation module was identified under a scale-free network with topological properties. Additionally, key driver analysis (KDA) was applied to the mRNA-lncRNA regulation module to identify the driver genes. The KDA revealed that ARD3 (FDR = 1.46 × 10-22), SND1 (FDR = 5.24 × 10-8), and lnc-DHX32-2:1 (FDR = 1.43 × 10-17) were particularly highlighted in the category of G2/M, G1/S, and M phases. Moreover, several genes we identified exhibited great connectivity in our causal gene network with every key driver gene, including CDK14, POLA1, lnc-NCS1-2:1, and lnc-FOXK1-4:1 (all FDR < 0.05 in those phases). Together, these results obtained using mathematical approaches and bioinformatics algorithmics might provide potential new mechanisms involved in the cytotoxicity induced by Cr.

Differential DNA Methylation of Networked Signaling, Transcriptional, Innate and Adaptive Immunity, and Osteoclastogenesis Genes and Pathways in Gout.

OBJECTIVE: In gout, autoinflammatory responses to urate crystals promote acute arthritis flares, but the pathogeneses of tophi, chronic synovitis, and erosion are less well understood. Defining the pathways of epigenomic immunity training can reveal novel pathogenetic factors and biomarkers. The present study was undertaken to seminally probe differential DNA methylation patterns utilizing epigenome-wide analyses in patients with gout. METHODS: Peripheral blood mononuclear cells (PBMCs) were obtained from a San Diego cohort of patients with gout (n = 16) and individually matched healthy controls (n = 14). PBMC methylome data were processed with ChAMP package in R. ENCODE data and Taiji data analysis software were used to analyze transcription factor (TF)-gene networks. As an independent validation cohort, whole blood DNA samples from New Zealand Maori subjects (n = 13 patients with gout, n = 16 control subjects without gout) were analyzed. RESULTS: Differentially methylated loci clearly separated gout patients from controls, as determined by hierarchical clustering and principal components analyses. IL23R, which mediates granuloma formation and cell invasion, was identified as one of the multiple differentially methylated gout risk genes. Epigenome-wide analyses revealed differential methylome pathway enrichment for B and T cell receptor signaling, Th17 cell differentiation and interleukin-17 signaling, convergent longevity regulation, circadian entrainment, and AMP-activated protein kinase signaling, which are pathways that impact inflammation via insulin-like growth factor 1 receptor, phosphatidylinositol 3-kinase/Akt, NF-κB, mechanistic target of rapamycin signaling, and autophagy. The gout cohorts overlapped for 37 (52.9%) of the 70 TFs with hypomethylated sequence enrichment and for 30 (78.9%) of the 38 enriched KEGG pathways identified via TFs. Evidence of shared differentially methylated gout TF-gene networks, including the NF-κB activation-limiting TFs MEF2C and NFATC2, pointed to osteoclast differentiation as the most strongly weighted differentially methylated pathway that overlapped in both gout cohorts. CONCLUSION: These findings of differential DNA methylation of networked signaling, transcriptional, innate and adaptive immunity, and osteoclastogenesis genes and pathways suggest that they could serve as novel therapeutic targets in the management of flares, tophi, chronic synovitis, and bone erosion in patients with gout.

VCNet: vector-based gene co-expression network construction and its application to RNA-seq data.

MOTIVATION: Building gene co-expression network (GCN) from gene expression data is an important field of bioinformatic research. Nowadays, RNA-seq data provides high dimensional information to quantify gene expressions in term of read counts for individual exons of genes. Such an increase in the dimension of expression data during the transition from microarray to RNA-seq era made many previous co-expression analysis algorithms based on simple univariate correlation no longer applicable. Recently, two vector-based methods, SpliceNet and RNASeqNet, have been proposed to build GCN. However, they failed to work when sample size is less than the number of exons. RESULTS: We develop an algorithm called VCNet to construct GCN from RNA-seq data to overcome this dimensional problem. VCNet performs a new statistical hypothesis test based on the correlation matrix of a gene-gene pair using the Frobenius norm. The asymptotic distribution of the new test is obtained under the null model. Simulation studies demonstrate that VCNet outperforms SpliceNet and RNASeqNet for detecting edges of GCN. We also apply VCNet to two expression datasets from TCGA database: the normal breast tissue and kidney tumour tissue, and the results show that the GCNs constructed by VCNet contain more biologically meaningful interactions than existing methods. CONCLUSION: VCNet is a useful tool to construct co-expression network. AVAILABILITY AND IMPLEMENTATION: VCNet is open source and freely available from https://github.com/wangzengmiao/VCNet under GNU LGPL v3. CONTACT: dengmh@pku.edu.cn or nelsontang@cuhk.edu.hk. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.