Mining Real-World Big Data to Characterize Adverse Drug Reaction Quantitatively: Mixed Methods Study.

BACKGROUND: Adverse drug reactions (ADRs), which are the phenotypic manifestations of clinical drug toxicity in humans, are a major concern in precision clinical medicine. A comprehensive evaluation of ADRs is helpful for unbiased supervision of marketed drugs and for discovering new drugs with high success rates. OBJECTIVE: In current practice, drug safety evaluation is often oversimplified to the occurrence or nonoccurrence of ADRs. Given the limitations of current qualitative methods, there is an urgent need for a quantitative evaluation model to improve pharmacovigilance and the accurate assessment of drug safety. METHODS: In this study, we developed a mathematical model, namely the Adverse Drug Reaction Classification System (ADReCS) severity-grading model, for the quantitative characterization of ADR severity, a crucial feature for evaluating the impact of ADRs on human health. The model was constructed by mining millions of real-world historical adverse drug event reports. A new parameter called Severity_score was introduced to measure the severity of ADRs, and upper and lower score boundaries were determined for 5 severity grades. RESULTS: The ADReCS severity-grading model exhibited excellent consistency (99.22%) with the expert-grading system, the Common Terminology Criteria for Adverse Events. Hence, we graded the severity of 6277 standard ADRs for 129,407 drug-ADR pairs. Moreover, we calculated the occurrence rates of 6272 distinct ADRs for 127,763 drug-ADR pairs in large patient populations by mining real-world medication prescriptions. With the quantitative features, we demonstrated example applications in systematically elucidating ADR mechanisms and thereby discovered a list of drugs with improper dosages. CONCLUSIONS: In summary, this study represents the first comprehensive determination of both ADR severity grades and ADR frequencies. This endeavor establishes a strong foundation for future artificial intelligence applications in discovering new drugs with high efficacy and low toxicity. It also heralds a paradigm shift in clinical toxicity research, moving from qualitative description to quantitative evaluation.

A distinct class of pan-cancer susceptibility genes revealed by an alternative polyadenylation transcriptome-wide association study.

Alternative polyadenylation plays an important role in cancer initiation and progression; however, current transcriptome-wide association studies mostly ignore alternative polyadenylation when identifying putative cancer susceptibility genes. Here, we perform a pan-cancer 3' untranslated region alternative polyadenylation transcriptome-wide association analysis by integrating 55 well-powered (n > 50,000) genome-wide association studies datasets across 22 major cancer types with alternative polyadenylation quantification from 23,955 RNA sequencing samples across 7,574 individuals. We find that genetic variants associated with alternative polyadenylation are co-localized with 28.57% of cancer loci and contribute a significant portion of cancer heritability. We further identify 642 significant cancer susceptibility genes predicted to modulate cancer risk via alternative polyadenylation, 62.46% of which have been overlooked by traditional expression- and splicing- studies. As proof of principle validation, we show that alternative alleles facilitate 3' untranslated region lengthening of CRLS1 gene leading to increased protein abundance and promoted proliferation of breast cancer cells. Together, our study highlights the significant role of alternative polyadenylation in discovering new cancer susceptibility genes and provides a strong foundational framework for enhancing our understanding of the etiology underlying human cancers.

scQTLbase: an integrated human single-cell eQTL database.

Genome-wide association studies (GWAS) have identified numerous genetic variants associated with diseases and traits. However, the functional interpretation of these variants remains challenging. Expression quantitative trait loci (eQTLs) have been widely used to identify mutations linked to disease, yet they explain only 20-50% of disease-related variants. Single-cell eQTLs (sc-eQTLs) studies provide an immense opportunity to identify new disease risk genes with expanded eQTL scales and transcriptional regulation at a much finer resolution. However, there is no comprehensive database dedicated to single-cell eQTLs that users can use to search, analyse and visualize them. Therefore, we developed the scQTLbase (http://bioinfo.szbl.ac.cn/scQTLbase), the first integrated human sc-eQTLs portal, featuring 304 datasets spanning 57 cell types and 95 cell states. It contains ∼16 million SNPs significantly associated with cell-type/state gene expression and ∼0.69 million disease-associated sc-eQTLs from 3 333 traits/diseases. In addition, scQTLbase offers sc-eQTL search, gene expression visualization in UMAP plots, a genome browser, and colocalization visualization based on the GWAS dataset of interest. scQTLbase provides a one-stop portal for sc-eQTLs that will significantly advance the discovery of disease susceptibility genes.

xQTLbiolinks: a comprehensive and scalable tool for integrative analysis of molecular QTLs.

Genome-wide association studies (GWAS) have identified thousands of disease-associated non-coding variants, posing urgent needs for functional interpretation. Molecular Quantitative Trait Loci (xQTLs) such as eQTLs serve as an essential intermediate link between these non-coding variants and disease phenotypes and have been widely used to discover disease-risk genes from many population-scale studies. However, mining and analyzing the xQTLs data presents several significant bioinformatics challenges, particularly when it comes to integration with GWAS data. Here, we developed xQTLbiolinks as the first comprehensive and scalable tool for bulk and single-cell xQTLs data retrieval, quality control and pre-processing from public repositories and our integrated resource. In addition, xQTLbiolinks provided a robust colocalization module through integration with GWAS summary statistics. The result generated by xQTLbiolinks can be flexibly visualized or stored in standard R objects that can easily be integrated with other R packages and custom pipelines. We applied xQTLbiolinks to cancer GWAS summary statistics as case studies and demonstrated its robust utility and reproducibility. xQTLbiolinks will profoundly accelerate the interpretation of disease-associated variants, thus promoting a better understanding of disease etiologies. xQTLbiolinks is available at https://github.com/lilab-bioinfo/xQTLbiolinks.

Immune-response 3'UTR alternative polyadenylation quantitative trait loci contribute to variation in human complex traits and diseases.

Genome-wide association studies (GWASs) have identified thousands of non-coding variants that are associated with human complex traits and diseases. The analysis of such GWAS variants in different contexts and physiological states is essential for deciphering the regulatory mechanisms underlying human disease. Alternative polyadenylation (APA) is a key post-transcriptional modification for most human genes that substantially impacts upon cell behavior. Here, we mapped 9,493 3'-untranslated region APA quantitative trait loci in 18 human immune baseline cell types and 8 stimulation conditions (immune 3'aQTLs). Through the comparison between baseline and stimulation data, we observed the high responsiveness of 3'aQTLs to immune stimulation (response 3'aQTLs). Co-localization and mendelian randomization analyses of immune 3'aQTLs identified 678 genes where 3'aQTL are associated with variation in complex traits, 27.3% of which were derived from response 3'aQTLs. Overall, these analyses reveal the role of immune 3'aQTLs in the determination of complex traits, providing new insights into the regulatory mechanisms underlying disease etiologies.

ipaQTL-atlas: an atlas of intronic polyadenylation quantitative trait loci across human tissues.

Functional interpretation of disease-associated non-coding variants remains a significant challenge in the post-GWAS era. Our recent study has identified 3'UTR alternative polyadenylation (APA) quantitative trait loci (3'aQTLs) and connects APA events with QTLs as a major driver of human traits and diseases. Besides 3'UTR, APA events can also occur in intron regions, and increasing evidence has connected intronic polyadenylation with disease risk. However, systematic investigation of the roles of intronic polyadenylation in human diseases remained challenging due to the lack of a comprehensive database across a variety of human tissues. Here, we developed ipaQTL-atlas (http://bioinfo.szbl.ac.cn/ipaQTL) as the first comprehensive portal for intronic polyadenylation. The ipaQTL-atlas is based on the analysis of 15 170 RNA-seq data from 838 individuals across 49 Genotype-Tissue Expression (GTEx v8) tissues and contains ∼0.98 million SNPs associated with intronic APA events. It provides an interface for ipaQTLs search, genome browser, boxplots, and data download, as well as the visualization of GWAS and ipaQTL colocalization results. ipaQTL-atlas provides a one-stop portal to access intronic polyadenylation information and could significantly advance the discovery of APA-associated disease susceptibility genes.

Using population-scale transcriptomic and genomic data to map 3' UTR alternative polyadenylation quantitative trait loci.

3' UTR alternative polyadenylation (APA) quantitative trait loci (3'aQTL) can explain approximately 16.1% of trait-associated non-coding variants and is largely distinct from other molecular QTLs. Here, we describe a bioinformatic protocol for identifying 3'aQTLs through standard RNA-seq and matched genomic data. This protocol allows users to analyze dynamic APA events, identify common genetic variants associated with differential 3' UTR usage, and predict the potential causal variants that affect APA. For complete details on the use and execution of this protocol, please refer to Li et al. (2021).

Discovering Innate Driver Variants for Risk Assessment of Early Colorectal Cancer Metastasis.

Metastasis is the main fatal cause of colorectal cancer (CRC). Although enormous efforts have been made to date to identify biomarkers associated with metastasis, there is still a huge gap to translate these efforts into effective clinical applications due to the poor consistency of biomarkers in dealing with the genetic heterogeneity of CRCs. In this study, a small cohort of eight CRC patients was recruited, from whom we collected cancer, paracancer, and normal tissues simultaneously and performed whole-exome sequencing. Given the exomes, a novel statistical parameter LIP was introduced to quantitatively measure the local invasion power for every somatic and germline mutation, whereby we affirmed that the innate germline mutations instead of somatic mutations might serve as the major driving force in promoting local invasion. Furthermore, via bioinformatic analyses of big data derived from the public zone, we identified ten potential driver variants that likely urged the local invasion of tumor cells into nearby tissue. Of them, six corresponding genes were new to CRC metastasis. In addition, a metastasis resister variant was also identified. Based on these eleven variants, we constructed a logistic regression model for rapid risk assessment of early metastasis, which was also deployed as an online server, AmetaRisk (http://www.bio-add.org/AmetaRisk). In summary, we made a valuable attempt in this study to exome-wide explore the genetic driving force to local invasion, which provides new insights into the mechanistic understanding of metastasis. Furthermore, the risk assessment model can assist in prioritizing therapeutic regimens in clinics and discovering new drug targets, and thus substantially increase the survival rate of CRC patients.

Determination of the Amino Acid Recruitment Order in Early Life by Genome-Wide Analysis of Amino Acid Usage Bias.

The mechanisms shaping the amino acids recruitment pattern into the proteins in the early life history presently remains a huge mystery. In this study, we conducted genome-wide analyses of amino acids usage and genetic codons structure in 7270 species across three domains of life. The carried-out analyses evidenced ubiquitous usage bias of amino acids that were likely independent from codon usage bias. Taking advantage of codon usage bias, we performed pseudotime analysis to re-determine the chronological order of the species emergence, which inspired a new species relationship by tracing the imprint of codon usage evolution. Furthermore, the multidimensional data integration showed that the amino acids A, D, E, G, L, P, R, S, T and V might be the first recruited into the last universal common ancestry (LUCA) proteins. The data analysis also indicated that the remaining amino acids most probably were gradually incorporated into proteogenesis process in the course of two long-timescale parallel evolutionary routes: I→F→Y→C→M→W and K→N→Q→H. This study provides new insight into the origin of life, particularly in terms of the basic protein composition of early life. Our work provides crucial information that will help in a further understanding of protein structure and function in relation to their evolutionary history.

Gene network analyses unveil possible molecular basis underlying drug-induced glaucoma.

BACKGROUND: Drug-induced glaucoma (DIG) is a kind of serious adverse drug reaction that can cause irreversible blindness. Up-to-date, the molecular mechanism of DIG largely remains unclear yet due to the medical complexity of glaucoma onset. METHODS: In this study, we conducted data mining of tremendous historical adverse drug events and genome-wide drug-regulated gene signatures to identify glaucoma-associated drugs. Upon these drugs, we carried out serial network analyses, including the weighted gene co-expression network analysis (WGCNA), to illustrate the gene interaction network underlying DIG. Furthermore, we applied pathogenic risk assessment to discover potential biomarker genes for DIG. RESULTS: As the results, we discovered 13 highly glaucoma-associated drugs, a glaucoma-related gene network, and 55 glaucoma-susceptible genes. These genes likely played central roles in triggering DIGs via an integrative mechanism of phototransduction dysfunction, intracellular calcium homeostasis disruption, and retinal ganglion cell death. Further pathogenic risk analysis manifested that a panel of nine genes, particularly OTOF gene, could serve as potential biomarkers for early-onset DIG prognosis. CONCLUSIONS: This study elucidates the possible molecular basis underlying DIGs systematically for the first time. It also provides prognosis clues for early-onset glaucoma and thus assists in designing better therapeutic regimens.

Broad-Spectrum Profiling of Drug Safety via Learning Complex Network.

Drug safety is a severe clinical pharmacology and toxicology problem that has caused immense medical and social burdens every year. Regretfully, a reproducible method to assess drug safety systematically and quantitatively is still missing. In this study, we developed an advanced machine learning model for de novo drug safety assessment by solving the multilayer drug-gene-adverse drug reaction (ADR) interaction network. For the first time, the drug safety was assessed in a broad landscape of 1,156 distinct ADRs. We also designed a parameter ToxicityScore to quantify the overall drug safety. Moreover, we determined association strength for every 3,807,631 gene-ADR interactions, which clues mechanistic exploration of ADRs. For convenience, we deployed the model as a web service ADRAlert-gene at http://www.bio-add.org/ADRAlert/. In summary, this study offers insights into prioritizing safe drug therapy. It helps reduce the attrition rate of new drug discovery by providing a reliable ADR profile in the early preclinical stage.

Down-regulation expression of TGFB2-AS1 inhibits the proliferation, migration, invasion and induces apoptosis in HepG2 cells.

BACKGROUND: Hepatocellular carcinoma (HCC) is the leading cause of cancer mortality and without effective prognosis. Previous study has been confirmed that the abnormal expression of long non-coding RNAs (lncRNAs) TGFB2-AS1 was involved in tumorigenesis. However, the biological functions of TGFB2-AS1 in hepatocellular carcinoma (HCC) remain largely unclear. OBJECTIVE: We comprehensively assess the clinical significance of TGFB2-AS1 and investigate the biological functions of TGFB2-AS1 on HCC HepG2 cells. METHODS: We firstly confirmed the expression of TGFB2-AS1 between tumor and normal tissues using public available transcriptome data. We analyzed the clinical significance of TGFB2-AS1 using the TCGA HCC datasets. The biological functions of TGFB2-AS1 on HCC HepG2 cells were explored by multiple in vitro assays. RESULTS: We found that TGFB2-AS1 was remarkably increased in HCC tissues (P = 0.00148) and exhibited a potential predictive marker for HCC, with an area under curve (AUC) of 0.708 (P = 0.0034) using the fifty pairs of matched HCC tissues of TCGA. Besides, higher expression of TGFB2-AS1 in HCC tissues was identified as being positively associated with advanced tumor (P = 0.012) and disease stage (P = 0.009) in 355 HCC cases using independent sample nonparametric test. Downregulation of TGFB2-AS1 expression significantly restrained proliferation (P < 0.01) and impaired colony formation (P < 0.05). Furthermore, TGFB2-AS1 depletion remarkably promoted the apoptosis of HepG2 cells (P < 0.05) and inhibited migration and invasion (P < 0.01). CONCLUSION: Taken together, these findings suggested that TGFB2-AS1 might serve as a potential therapeutic target for HCC.

EnhancerDB: a resource of transcriptional regulation in the context of enhancers.

Enhancers can act as cis-regulatory elements to control transcriptional regulation by recruiting DNA-binding transcription factors (TFs) in a tissue-specific manner. Recent studies show that enhancers regulate not only protein-coding genes but also microRNAs (miRNAs), and mutations within the TF binding sites (TFBSs) located on enhancers will cause a variety of diseases such as cancer. However, a comprehensive resource to integrate these regulation elements for revealing transcriptional regulations in the context of enhancers is not currently available. Here, we introduce EnhancerDB, a web-accessible database to provide a resource to browse and search regulatory relationships identified in this study, including 131 054 581 TF-enhancer, 17 059 enhancer-miRNAs, 318 993 enhancer-genes, 4 639 558 TF-miRNAs, 1 059 695 TF-genes, 11 439 394 enhancer-single-nucleotide polymorphisms (SNPs) and 23 334 genes associated with expression quantitative trait loci (eQTL) SNP and expression profile of TF/gene/miRNA across multiple human tissues/cell lines. We also developed a tool that further allows users to define tissue-specific enhancers by setting the threshold score of tissue specificity of enhancers. In addition, links to external resources are also available at EnhancerDB.

Genome-wide Identification and Characterization of Enhancers Across 10 Human Tissues.

Background: Enhancers can act as cis-regulatory elements (CREs) to control development and cellular function by regulating gene expression in a tissue-specific and ubiquitous manner. However, the regulatory network and characteristic of different types of enhancers (e.g., transcribed/non-transcribed enhancers, tissue-specific/ubiquitous enhancers) across multiple tissues are still unclear. Results: Here, a total of 53,924 active enhancers and 10,307 enhancer-associated RNAs (eRNAs) in 10 tissues (adrenal, brain, breast, heart, liver, lung, ovary, placenta, skeletal muscle and kidney) were identified through the integration of histone modifications (H3K4me1, H3K27ac and H3K4me3) and DNase I hypersensitive sites (DHSs) data. Moreover, 40,101 tissue-specific enhancers (TS-Enh), 1,241 ubiquitously expressed enhancers (UE-Enh) as well as transcribed enhancers (T-Enh), including 7,727 unidirectionally transcribed enhancers (1D-Enh) and 1,215 bidirectionally transcribed enhancers (2D-Enh) were defined in 10 tissues. The results show that enhancers exhibited high GC content, genomic variants and transcription factor binding sites (TFBS) enrichment in all tissues. These characteristics were significantly different between TS-Enh and UE-Enh, T-Enh and NT-Enh, 2D-Enh and 1D-Enh. Furt hermore, the results showed that enhancers obviously upregulate the expression of adjacent target genes which were remarkably correlated with the functions of corresponding tissues. Finally, a free user-friendly tissue-specific enhancer database, TiED (http://lcbb.swjtu.edu.cn/TiED), has been built to store, visualize, and confer these results. Conclusion: Genome-wide analysis of the regulatory network and characteristic of various types of enhancers showed that enhancers associated with TFs, eRNAs and target genes appeared in tissue specificity and function across different tissues.

Identification and Analysis of P53-Mediated Competing Endogenous RNA Network in Human Hepatocellular Carcinoma.

Recent studies have indicated that long non-coding RNAs (lncRNAs) and mRNA function as competing endogenous RNAs (ceRNAs) that compete to bind to shared microRNA (miRNA) recognition elements (MREs) to perform specific biological functions during tumorigenesis. The tumor suppressor p53 is a master regulator of cancer-related biological processes by acting as a transcription factor to regulate target genes including miRNA and lncRNA. However, the mechanism in human hepatocellular carcinoma and whether p53-mediated RNA targets could form ceRNA network remain unclear. Here, we identified a series of differential expressed miRNAs, lncRNA and mRNA which were potentially regulated by p53 using RNA sequencing in HepG2. Genomic characteristics comparative analysis showed significant differences between mRNAs and lncRNAs. By integrating experimentally confirmed Ago2 and p53 binding sites, we constructed a highly reliable p53-mediated ceRNA network using hypergeometric test. The KEGG pathway enrichment analysis showed that the ceRNA network highly enriched in the cancer or p53-associated signaling pathways. Finally, using betweenness centrality analysis, we identified five master miRNAs (hsa-miR-3620-5p, hsa-miR-3613-3p, hsa-miR-6881-3p, hsa-miR-6087 and hsa-miR-18a-3p) that regulated most of the target RNAs, suggesting these miRNAs play central roles in the whole p53-mediated ceRNAs network. Taken together, our results provide a new regulatory mechanism of p53 networks for future studies in cancer therapeutics.

Transcriptome profiling analysis of differentially expressed mRNAs and lncRNAs in HepG2 cells treated with peptide 9R-P201.

OBJECTIVE: To characterize the transcriptome profile of hepatocellular carcinoma (HCC) HepG2 cells treated with peptide 9R-P201 for further functional verification and HCC drug development. RESULTS: 1557 mRNAs (1125 upregulated and 432 downregulated) and 881 lncRNAs (640 upregulated and 241 downregulated) with significant differential expression were identified using RNA-seq. The qRT-PCR results showed that the differential expression of several mRNAs and lncRNAs coincided with the RNA-seq results. Differentially expressed mRNAs and lncRNAs presented a significant difference in genomic characteristics but no preference under 9R-P201 treatment compared with control. The GO and KEGG functional enrichment analyses showed that differentially expressed mRNAs and lncRNAs remarkably enriched in cancer-related biological processes and signaling pathways. Finally, we screened out 33 TFs, 273 lncRNAs and 94 target genes with high degree interaction which were remarkably associated with the tumorigenesis and progression of cancers using betweenness centrality analysis. CONCLUSION: These findings offer novel insights into the mechanism of 9R-P201 in HepG2 cells and provide new opportunities for the future 9R-P201-based drug development and the treatment of hepatocellular carcinoma.

Comprehensive Expression Profiling and Functional Network Analysis of p53-Regulated MicroRNAs in HepG2 Cells Treated with Doxorubicin.

Acting as a sequence-specific transcription factor, p53 tumor suppressor involves in a variety of biological processes after being activated by cellular stresses such as DNA damage. In recent years, microRNAs (miRNAs) have been confirmed to be regulated by p53 in several cancer types. However, it is still unclear how miRNAs orchestrate their regulation and function in p53 network after p53 activation in hepatocellular carcinoma (HCC). In this study, we used small RNA sequencing and systematic bioinformatic analysis to characterize the regulatory networks of differentially expressed miRNAs after the p53 activation in HepG2. Here, 33 miRNAs significantly regulated by p53 (12 up-regulated and 21 down-regulated) were detected between the doxorubicin-treated and untreated HepG2 cells in two biological replicates for small RNA sequencing and 8 miRNAs have been reported previously to be associated with HCC. Gene ontology (GO) and KEGG pathway enrichment analysis showed that 87.9% (29 out of 33) and 90.9% (30 out of 33) p53-regulated miRNAs were involved in p53-related biological processes and pathways with significantly low p-value, respectively. Remarkably, 18 out of 33 p53-regulated miRNAs were identified to contain p53 binding sites around their transcription start sites (TSSs). Finally, comprehensive p53-miRNA regulatory networks were constructed and analyzed. These observations provide a new insight into p53-miRNA co-regulatory network in the context of HCC.

Genome-wide survey of tissue-specific microRNA and transcription factor regulatory networks in 12 tissues.

Tissue-specific miRNAs (TS miRNA) specifically expressed in particular tissues play an important role in tissue identity, differentiation and function. However, transcription factor (TF) and TS miRNA regulatory networks across multiple tissues have not been systematically studied. Here, we manually extracted 116 TS miRNAs and systematically investigated the regulatory network of TF-TS miRNA in 12 human tissues. We identified 2,347 TF-TS miRNA regulatory relations and revealed that most TF binding sites tend to enrich close to the transcription start site of TS miRNAs. Furthermore, we found TS miRNAs were regulated widely by non-tissue specific TFs and the tissue-specific expression level of TF have a close relationship with TF-genes regulation. Finally, we describe TSmiR (http://bioeng.swjtu.edu.cn/TSmiR), a novel and web-searchable database that houses interaction maps of TF-TS miRNA in 12 tissues. Taken together, these observations provide a new suggestion to better understand the regulatory network and mechanisms of TF-TS miRNAs underlying different tissues.