Bioinformatics Research on Drug Sensitivity Prediction.

Modeling-based anti-cancer drug sensitivity prediction has been extensively studied in recent years. While most drug sensitivity prediction models only use gene expression data, the remarkable impacts of gene mutation, methylation, and copy number variation on drug sensitivity are neglected. Drug sensitivity prediction can both help protect patients from some adverse drug reactions and improve the efficacy of treatment. Genomics data are extremely useful for drug sensitivity prediction task. This article reviews the role of drug sensitivity prediction, describes a variety of methods for predicting drug sensitivity. Moreover, the research significance of drug sensitivity prediction, as well as existing problems are well discussed.

SC2disease: a manually curated database of single-cell transcriptome for human diseases.

SC2disease (http://easybioai.com/sc2disease/) is a manually curated database that aims to provide a comprehensive and accurate resource of gene expression profiles in various cell types for different diseases. With the development of single-cell RNA sequencing (scRNA-seq) technologies, uncovering cellular heterogeneity of different tissues for different diseases has become feasible by profiling transcriptomes across cell types at the cellular level. In particular, comparing gene expression profiles between different cell types and identifying cell-type-specific genes in various diseases offers new possibilities to address biological and medical questions. However, systematic, hierarchical and vast databases of gene expression profiles in human diseases at the cellular level are lacking. Thus, we reviewed the literature prior to March 2020 for studies which used scRNA-seq to study diseases with human samples, and developed the SC2disease database to summarize all the data by different diseases, tissues and cell types. SC2disease documents 946 481 entries, corresponding to 341 cell types, 29 tissues and 25 diseases. Each entry in the SC2disease database contains comparisons of differentially expressed genes between different cell types, tissues and disease-related health status. Furthermore, we reanalyzed gene expression matrix by unified pipeline to improve the comparability between different studies. For each disease, we also compare cell-type-specific genes with the corresponding genes of lead single nucleotide polymorphisms (SNPs) identified in genome-wide association studies (GWAS) to implicate cell type specificity of the traits.

Comprehensive Analysis of Differentially Expressed lncRNAs in Gastric Cancer.

Gastric cancer (GC) is the fourth most common malignant tumor. The mechanism underlying GC occurrence and development remains unclear. Previous studies have indicated that long non-coding RNAs (lncRNAs) are significantly associated with gastric cancer, but a systematic understanding of the role of lncRNAs in gastric cancer is lacking. In recent years, with the development of next-generation sequencing technology, tens of thousands of lncRNAs have been discovered. However, a large number of unannotated lncRNAs remain unidentified in different tissues, including potential gastric cancer-related lncRNAs. In this study, RNA sequencing (RNA-seq) data from 16 samples of eight gastric cancer patients were obtained and analyzed. A total of 1,854 previously unannotated lncRNAs were identified by ab initio assembly, and 520 differentially expressed lncRNAs were validated in the TCGA expression dataset. Methylation and copy number variation (CNV) array data from the same sample were integrated in the analysis. Changes in DNA methylation levels and CNVs may be responsible for the differential expression of 91 lncRNAs. Differentially expressed lncRNAs were enriched in coexpressed clusters of genes related to functions such as cell signaling, cell cycle, immune response, metabolic processes, angiogenesis, and regulation of retinoic acid (RA) receptors. Finally, a differentially expressed lncRNA, AC004510.3, was identified as a potential biomarker for the prediction of the overall survival of gastric cancer patients.

LncDisAP: a computation model for LncRNA-disease association prediction based on multiple biological datasets.

BACKGROUND: Over the past decades, a large number of long non-coding RNAs (lncRNAs) have been identified. Growing evidence has indicated that the mutation and dysregulation of lncRNAs play a critical role in the development of many complex human diseases. Consequently, identifying potential disease-related lncRNAs is an effective means to improve the quality of disease diagnostics and treatment, which is the motivation of this work. Here, we propose a computational model (LncDisAP) for potential disease-related lncRNA identification based on multiple biological datasets. First, the associations between lncRNA and different data sources are collected from different databases. With these data sources as dimensions, we calculate the functional associations between lncRNAs by the recommendation strategy of collaborative filtering. Subsequently, a disease-associated lncRNA functional network is built with functional similarities between lncRNAs as the weight. Ultimately, potential disease-related lncRNAs can be identified based on ranked scores derived by random walking with restart (RWR). Then, training sets and testing sets are extracted from two different versions of a disease-lncRNA dataset to assess the performance of LncDisAP on 54 diseases. RESULTS: A lncRNA functional network is built based on the proposed computational model, and it contains 66,060 associations among 364 lncRNAs associated with 182 diseases in total. We extract 218 known disease-lncRNA pairs associated with 54 diseases to assess the network. As a result, the average AUC (area under the receiver operating characteristic curve) of LncDisAP is 78.08%. CONCLUSION: In this article, a computational model integrating multiple lncRNA-related biological datasets is proposed for identifying potential disease-related lncRNAs. The result shows that LncDisAP is successful in predicting novel disease-related lncRNA signatures. In addition, with several common cancers taken as case studies, we found some unknown lncRNAs that could be associated with these diseases through our network. These results suggest that this method can be helpful in improving the quality for disease diagnostics and treatment.

Estrogen induces global reorganization of chromatin structure in human breast cancer cells.

In the cell nucleus, each chromosome is confined to a chromosome territory. This spatial organization of chromosomes plays a crucial role in gene regulation and genome stability. An additional level of organization has been discovered at the chromosome scale: the spatial segregation into open and closed chromatins to form two genome-wide compartments. Although considerable progress has been made in our knowledge of chromatin organization, a fundamental issue remains the understanding of its dynamics, especially in cancer. To address this issue, we performed genome-wide mapping of chromatin interactions (Hi-C) over the time after estrogen stimulation of breast cancer cells. To biologically interpret these interactions, we integrated with estrogen receptor α (ERα) binding events, gene expression and epigenetic marks. We show that gene-rich chromosomes as well as areas of open and highly transcribed chromatins are rearranged to greater spatial proximity, thus enabling genes to share transcriptional machinery and regulatory elements. At a smaller scale, differentially interacting loci are enriched for cancer proliferation and estrogen-related genes. Moreover, these loci are correlated with higher ERα binding events and gene expression. Taken together these results reveal the role of a hormone--estrogen--on genome organization, and its effect on gene regulation in cancer.

The personal genome browser: visualizing functions of genetic variants.

Advances in high-throughput sequencing technologies have brought us into the individual genome era. Projects such as the 1000 Genomes Project have led the individual genome sequencing to become more and more popular. How to visualize, analyse and annotate individual genomes with knowledge bases to support genome studies and personalized healthcare is still a big challenge. The Personal Genome Browser (PGB) is developed to provide comprehensive functional annotation and visualization for individual genomes based on the genetic-molecular-phenotypic model. Investigators can easily view individual genetic variants, such as single nucleotide variants (SNVs), INDELs and structural variations (SVs), as well as genomic features and phenotypes associated to the individual genetic variants. The PGB especially highlights potential functional variants using the PGB built-in method or SIFT/PolyPhen2 scores. Moreover, the functional risks of genes could be evaluated by scanning individual genetic variants on the whole genome, a chromosome, or a cytoband based on functional implications of the variants. Investigators can then navigate to high risk genes on the scanned individual genome. The PGB accepts Variant Call Format (VCF) and Genetic Variation Format (GVF) files as the input. The functional annotation of input individual genome variants can be visualized in real time by well-defined symbols and shapes. The PGB is available at http://www.pgbrowser.org/.

Amplification of distant estrogen response elements deregulates target genes associated with tamoxifen resistance in breast cancer.

A causal role of gene amplification in tumorigenesis is well known, whereas amplification of DNA regulatory elements as an oncogenic driver remains unclear. In this study, we integrated next-generation sequencing approaches to map distant estrogen response elements (DEREs) that remotely control the transcription of target genes through chromatin proximity. Two densely mapped DERE regions located on chromosomes 17q23 and 20q13 were frequently amplified in estrogen receptor-α-positive luminal breast cancer. These aberrantly amplified DEREs deregulated target gene expression potentially linked to cancer development and tamoxifen resistance. Progressive accumulation of DERE copies was observed in normal breast progenitor cells chronically exposed to estrogenic chemicals. These findings may extend to other DNA regulatory elements, the amplification of which can profoundly alter target transcriptome during tumorigenesis.

Identification of regulatory regions of bidirectional genes in cervical cancer.

BACKGROUND: Bidirectional promoters are shared promoter sequences between divergent gene pair (genes proximal to each other on opposite strands), and can regulate the genes in both directions. In the human genome, > 10% of protein-coding genes are arranged head-to-head on opposite strands, with transcription start sites that are separated by < 1,000 base pairs. Many transcription factor binding sites occur in the bidirectional promoters that influence the expression of 2 opposite genes. Recently, RNA polymerase II (RPol II) ChIP-seq data are used to identify the promoters of coding genes and non-coding RNAs. However, a bidirectional promoter with RPol II ChIP-Seq data has not been found. RESULTS: In some bidirectional promoter regions, the RPol II forms a bi-peak shape, which indicates that 2 promoters are located in the bidirectional region. We have developed a computational approach to identify the regulatory regions of all divergent gene pairs using genome-wide RPol II binding patterns derived from ChIP-seq data, based upon the assumption that the distribution of RPol II binding patterns around the bidirectional promoters are accumulated by RPol II binding of 2 promoters. In HeLa S3 cells, 249 promoter pairs and 1094 single promoters were identified, of which 76 promoters cover only positive genes, 86 promoters cover only negative genes, and 932 promoters cover 2 genes. Gene expression levels and STAT1 binding sites for different promoter categories were therefore examined. CONCLUSIONS: The regulatory region of bidirectional promoter identification based upon RPol II binding patterns provides important temporal and spatial measurements regarding the initiation of transcription. From gene expression and transcription factor binding site analysis, the promoters in bidirectional regions may regulate the closest gene, and STAT1 is involved in primary promoter.

Potential roles of microRNAs in regulating long intergenic noncoding RNAs.

BACKGROUND: Over 10,000 long intergenic non-coding RNAs (lincRNAs) have been identified in the human genome. Some have been well characterized and known to participate in various stages of gene regulation. In the post-transcriptional process, another class of well-known small non-coding RNA, or microRNA (miRNA), is very active in inhibiting mRNA. Though similar features between mRNA and lincRNA have been revealed in several recent studies, and a few isolated miRNA-lincRNA relationships have been observed. Despite these advances, the comprehensive miRNA regulation pattern of lincRNA has not been clarified. METHODS: In this study, we investigated the possible interaction between the two classes of non-coding RNAs. Instead of using the existing long non-coding database, we employed an ab initio method to annotate lincRNAs expressed in a group of normal breast tissues and breast tumors. RESULTS: Approximately 90 lincRNAs show strong reverse expression correlation with miRNAs, which have at least one predicted target site presented. These target sites are statistically more conserved than their neighboring genetic regions and other predicted target sites. Several miRNAs that target to these lincRNAs are known to play an essential role in breast cancer. CONCLUSION: Similar to inhibiting mRNAs, miRNAs show potential in promoting the degeneration of lincRNAs. Breast-cancer-related miRNAs may influence their target lincRNAs resulting in differential expression in normal and malignant breast tissues. This implies the miRNA regulation of lincRNAs may be involved in the regulatory process in tumor cells.

Prioritization of disease microRNAs through a human phenome-microRNAome network.

BACKGROUND: The identification of disease-related microRNAs is vital for understanding the pathogenesis of diseases at the molecular level, and is critical for designing specific molecular tools for diagnosis, treatment and prevention. Experimental identification of disease-related microRNAs poses considerable difficulties. Computational analysis of microRNA-disease associations is an important complementary means for prioritizing microRNAs for further experimental examination. RESULTS: Herein, we devised a computational model to infer potential microRNA-disease associations by prioritizing the entire human microRNAome for diseases of interest. We tested the model on 270 known experimentally verified microRNA-disease associations and achieved an area under the ROC curve of 75.80%. Moreover, we demonstrated that the model is applicable to diseases with which no known microRNAs are associated. The microRNAome-wide prioritization of microRNAs for 1,599 disease phenotypes is publicly released to facilitate future identification of disease-related microRNAs. CONCLUSIONS: We presented a network-based approach that can infer potential microRNA-disease associations and drive testable hypotheses for the experimental efforts to identify the roles of microRNAs in human diseases.