In silico pathway analysis and tissue specific cis-eQTL for colorectal cancer GWAS risk variants

Transcriptional dynamics of colorectal cancer risk associated variation at 11q23.1 correlate with tuft cell abundance and marker expression in silico

Colorectal cancer (CRC) is characterised by heritable risk that is not well understood. Heritable, genetic variation at 11q23.1 is associated with increased colorectal cancer (CRC) risk, demonstrating eQTL effects on 3 cis- and 23 trans-eQTL targets. We sought to determine the relationship between 11q23.1 cis- and trans-eQTL target expression and test for potential cell-specificity. scRNAseq from 32,361 healthy colonic epithelial cells was aggregated and subject to weighted gene co-expression network analysis (WGCNA). One module (blue) included 19 trans-eQTL targets and was correlated with POU2AF2 expression only. Following unsupervised clustering of single cells, the expression of 19 trans-eQTL targets was greatest and most variable in cluster number 11, which transcriptionally resembled tuft cells. 14 trans-eQTL targets were found to demarcate this cluster, 11 of which were corroborated in a second dataset. Intra-cluster WGCNA and module preservation analysis then identified twelve 11q23.1 trans-eQTL targets to comprise a network that was specific to cluster 11. Finally, linear modelling and differential abundance testing showed 11q23.1 trans-eQTL target expression was predictive of cluster 11 abundance. Our findings suggest 11q23.1 trans-eQTL targets comprise a POU2AF2-related network that is likely tuft cell-specific and reduced expression of these genes correlates with reduced tuft cell abundance in silico.

XGBG: A Novel Method for Identifying Ovarian Carcinoma Susceptible Genes Based on Deep Learning

Ovarian carcinomas (OCs) represent a heterogeneous group of neoplasms consisting of several entities with pathogenesis, molecular profiles, multiple risk factors, and outcomes. OC has been regarded as the most lethal cancer among women all around the world. There are at least five main types of OCs classified by the fifth edition of the World Health Organization of tumors: high-/low-grade serous carcinoma, mucinous carcinoma, clear cell carcinoma, and endometrioid carcinoma. With the improved knowledge of genome-wide association study (GWAS) and expression quantitative trait locus (eQTL) analyses, the knowledge of genomic landscape of complex diseases has been uncovered in large measure. Moreover, pathway analyses also play an important role in exploring the underlying mechanism of complex diseases by providing curated pathway models and information about molecular dynamics and cellular processes. To investigate OCs deeper, we introduced a novel disease susceptible gene prediction method, XGBG, which could be used in identifying OC-related genes based on different omics data and deep learning methods. We first employed the graph convolutional network (GCN) to reconstruct the gene features based on both gene feature and network topological structure. Then, a boosting method is utilized to predict OC susceptible genes. As a result, our model achieved a high AUC of 0.7541 and an AUPR of 0.8051, which indicates the effectiveness of the XGPG. Based on the newly predicted OC susceptible genes, we gathered and researched related literatures to provide strong support to the results, which may help in understanding the pathogenesis and mechanisms of the disease.

An Ovarian Cancer Susceptible Gene Prediction Method Based on Deep Learning Methods

Ovarian cancer (OC) is one of the most fatal diseases among women all around the world. It is highly lethal because it is usually diagnosed at an advanced stage which may reduce the survival rate greatly. Even though most of the patients are treated timely and effectively, the survival rate is still low due to the high recurrence rate of OC. With a large number of genome-wide association analysis (GWAS)-discovered risk regions of OC, expression quantitative trait locus (eQTL) analyses can explore candidate susceptible genes based on these risk loci. However, a large number of OC-related genes remain unknown. In this study, we proposed a novel gene prediction method based on different omics data and deep learning methods to identify OC causal genes. We first employed graph attention network (GAT) to obtain a compact gene feature representation, then a deep neural network (DNN) is utilized to predict OC-related genes. As a result, our model achieved a high AUC of 0.761 and AUPR of 0.788, which proved the accuracy and effectiveness of our proposed method. At last, we conducted a gene-set enrichment analysis to further explore the mechanism of OC. Finally, we predicted 245 novel OC causal genes and 10 top related KEGG pathways.

Differential genetic influences over colorectal cancer risk and gene expression in large bowel mucosa

Site-specific variation in colorectal cancer (CRC) incidence, biology and prognosis are poorly understood. We sought to determine whether common genetic variants influencing CRC risk might exhibit topographical differences on CRC risk through regional differences in effects on gene expression in the large bowel mucosa. We conducted a site-specific genetic association study (10 630 cases, 31 331 controls) to identify whether established risk variants exert differential effects on risk of proximal, compared to distal CRC. We collected normal colorectal mucosa and blood from 481 subjects and assessed mucosal gene expression using Illumina HumanHT-12v4 arrays in relation to germline genotype. Expression quantitative trait loci (eQTLs) were explored by anatomical location of sampling. The rs3087967 genotype (chr11q23.1 risk variant) exhibited significant site-specific effects-risk of distal CRC (odds ratio [OR] = 1.20, P = 8.20 × 10-20 ) with negligible effects on proximal CRC risk (OR = 1.05, P = .10). Expression of 1261 genes differed between proximal and distal colonic mucosa (top hit PRAC gene, fold-difference = 10, P = 3.48 × 10-57 ). In eQTL studies, rs3087967 genotype was associated with expression of 8 cis- and 21 trans-genes. Four of these (AKAP14, ADH5P4, ASGR2, RP11-342M1.7) showed differential effects by site, with strongest trans-eQTL signals in proximal colonic mucosa (eg, AKAP14, beta = 0.61, P = 5.02 × 10-5 ) and opposite signals in distal mucosa (AKAP14, beta = -0.17, P = .04). In summary, genetic variation at the chr11q23.1 risk locus imparts greater risk of distal rather than proximal CRC and exhibits site-specific differences in eQTL effects in normal mucosa. Topographical differences in genomic control over gene expression relevant to CRC risk may underlie site-specific variation in CRC. Results may inform individualised CRC screening programmes.

Associations between pancreatic expression quantitative traits and risk of pancreatic ductal adenocarcinoma

Pancreatic ductal adenocarcinoma (PDAC) is among the most lethal cancers. Its poor prognosis is predominantly due to the fact that most patients remain asymptomatic until the disease reaches an advanced stage, alongside the lack of early markers and screening strategies. A better understanding of PDAC risk factors is essential for the identification of groups at high risk in the population. Genome-wide association studies (GWAS) have been a powerful tool for detecting genetic variants associated with complex traits, including pancreatic cancer. By exploiting functional and GWAS data, we investigated the associations between polymorphisms affecting gene function in the pancreas (expression quantitative trait loci, eQTLs) and PDAC risk. In a two-phase approach, we analysed 13 713 PDAC cases and 43 784 controls and identified a genome-wide significant association between the A allele of the rs2035875 polymorphism and increased PDAC risk (P = 7.14 × 10-10). This allele is known to be associated with increased expression in the pancreas of the keratin genes KRT8 and KRT18, whose increased levels have been reported to correlate with various tumour cell characteristics. Additionally, the A allele of the rs789744 variant was associated with decreased risk of developing PDAC (P = 3.56 × 10-6). This single nucleotide polymorphism is situated in the SRGAP1 gene and the A allele is associated with higher expression of the gene, which in turn inactivates the cyclin-dependent protein 42 (CDC42) gene expression, thus decreasing the risk of PDAC. In conclusion, we present here a functional-based novel PDAC risk locus and an additional strong candidate supported by significant associations and plausible biological mechanisms.

Non-Lynch Familial and Early-Onset Colorectal Cancer Explained by Accumulation of Low-Risk Genetic Variants

A large proportion of familial and/or early-onset cancer patients do not carry pathogenic variants in known cancer predisposing genes. We aimed to assess the contribution of previously validated low-risk colorectal cancer (CRC) alleles to familial/early-onset CRC (fCRC) and to serrated polyposis. We estimated the association of CRC with a 92-variant-based weighted polygenic risk score (wPRS) using 417 fCRC patients, 80 serrated polyposis patients, 1077 hospital-based incident CRC patients, and 1642 controls. The mean wPRS was significantly higher in fCRC than in controls or sporadic CRC patients. fCRC patients in the highest (20th) wPRS quantile were at four-fold greater CRC risk than those in the middle quantile (10th). Compared to low-wPRS fCRC, a higher number of high-wPRS fCRC patients had developed multiple primary CRCs, had CRC family history, and were diagnosed at age ≥50. No association with wPRS was observed for serrated polyposis. In conclusion, a relevant proportion of mismatch repair (MMR)-proficient fCRC cases might be explained by the accumulation of low-risk CRC alleles. Validation in independent cohorts and development of predictive models that include polygenic risk score (PRS) data and other CRC predisposing factors will determine the implementation of PRS into genetic testing and counselling in familial and early-onset CRC.

Identifying Transcriptomic Signatures and Rules for SARS-CoV-2 Infection

The world-wide Coronavirus Disease 2019 (COVID-19) pandemic was triggered by the widespread of a new strain of coronavirus named as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Multiple studies on the pathogenesis of SARS-CoV-2 have been conducted immediately after the spread of the disease. However, the molecular pathogenesis of the virus and related diseases has still not been fully revealed. In this study, we attempted to identify new transcriptomic signatures as candidate diagnostic models for clinical testing or as therapeutic targets for vaccine design. Using the recently reported transcriptomics data of upper airway tissue with acute respiratory illnesses, we integrated multiple machine learning methods to identify effective qualitative biomarkers and quantitative rules for the distinction of SARS-CoV-2 infection from other infectious diseases. The transcriptomics data was first analyzed by Boruta so that important features were selected, which were further evaluated by the minimum redundancy maximum relevance method. A feature list was produced. This list was fed into the incremental feature selection, incorporating some classification algorithms, to extract qualitative biomarker genes and construct quantitative rules. Also, an efficient classifier was built to identify patients infected with SARS-COV-2. The findings reported in this study may help in revealing the potential pathogenic mechanisms of COVID-19 and finding new targets for vaccine design.

Two variants of Interleukin-1B gene are associated with the decreased risk, clinical features, and better overall survival of colorectal cancer: a two-center case-control study

Interleukin (IL)-1B reportedly promotes the stemness and invasiveness of colon cancer cells. Several studies have investigated the association between IL-1B gene polymorphisms and colorectal cancer (CRC) risk, but report conflicting findings. Here, this association was explored in a hospital-based case-control study involving 527 CRC cases and 639 controls from two Chinese Han populations. Genotyping was done by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. The IL-1B expression in CRC patients and controls were obtained from the GEPIA database and the mRNA expressions of different genotypes were downloaded from the GTEx portal database. The relationship of two IL-1B gene loci with clinical parameters and overall survival were analyzed using the Chi-square test and Kaplan-Meier analysis with the log-rank test respectively. It was found that the IL-1B mRNA levels in CRC patients were significantly higher than in controls. Bioinformatic analysis suggested that rs1143634 and rs1143623 polymorphisms decreased the IL-1B mRNA expression. The two polymorphisms were associated with decreased risk for CRC. Stratified analyses revealed the IL-1B gene rs1143623 and rs1143634 polymorphisms decreased the risk of CRC among females, smokers and drinkers. Moreover, the CC and/or GC genotype of rs1143623 polymorphism were correlated with decreased risk among CRC patients with tumor size ≥5cm, TNM stage III+IV, and rectal cancer. For rs1143634 polymorphism, the CT genotype reduced the risk of colorectal adenocarcinoma. The CRC patients carrying CC genotype of rs1143623 polymorphism were associated with better overall survival. In conclusion, IL-1B gene rs1143623 and rs1143634 polymorphisms are associated with decreased risk for CRC patients and thereby play important roles in the etiology of CRC.

Identifying Putative Susceptibility Genes and Evaluating Their Associations with Somatic Mutations in Human Cancers

Genome-wide association studies (GWASs) have identified hundreds of genetic risk variants for human cancers. However, target genes for the majority of risk loci remain largely unexplored. It is also unclear whether GWAS risk-loci-associated genes contribute to mutational signatures and tumor mutational burden (TMB) in cancer tissues. We systematically conducted cis-expression quantitative trait loci (cis-eQTL) analyses for 294 GWAS-identified variants for six major types of cancer-colorectal, lung, ovary, prostate, pancreas, and melanoma-by using transcriptome data from the Genotype-Tissue Expression (GTEx) Project, the Cancer Genome Atlas (TCGA), and other public data sources. By using integrative analysis strategies, we identified 270 candidate target genes, including 99 with previously unreported associations, for six cancer types. By analyzing functional genomic data, our results indicate that 180 genes (66.7% of 270) had evidence of cis-regulation by putative functional variants via proximal promoter or distal enhancer-promoter interactions. Together with our previously reported associations for breast cancer risk, our results show that 24 genes are shared by at least two cancer types, including four genes for both breast and ovarian cancer. By integrating mutation data from TCGA, we found that expression levels of 33 and 66 putative susceptibility genes were associated with specific mutational signatures and TMB of cancer-driver genes, respectively, at a Bonferroni-corrected p < 0.05. Together, these findings provide further insight into our understanding of how genetic risk variants might contribute to carcinogenesis through the regulation of susceptibility genes that are related to the biogenesis of somatic mutations.

SNP association study in PMS2-associated Lynch syndrome

Lynch syndrome (LS) patients are at high risk of developing colorectal cancer (CRC). Phenotypic variability might in part be explained by common susceptibility loci identified in Genome Wide Association Studies (GWAS). Previous studies focused mostly on MLH1, MSH2 and MSH6 carriers, with conflicting results. We aimed to determine the role of GWAS SNPs in PMS2 mutation carriers. A cohort study was performed in 507 PMS2 carriers (124 CRC cases), genotyped for 24 GWAS SNPs, including SNPs at 11q23.1 and 8q23.3. Hazard ratios (HRs) were calculated using a weighted Cox regression analysis to correct for ascertainment bias. Discrimination was assessed with a concordance statistic in a bootstrap cross-validation procedure. Individual SNPs only had non-significant associations with CRC occurrence with HRs lower than 2, although male carriers of allele A at rs1321311 (6p21.31) may have increased risk of CRC (HR = 2.1, 95% CI 1.2–3.0). A polygenic risk score (PRS) based on 24 HRs had an HR of 2.6 (95% CI 1.5–4.6) for the highest compared to the lowest quartile, but had no discriminative ability (c statistic 0.52). Previously suggested SNPs do not modify CRC risk in PMS2 carriers. Future large studies are needed for improved risk stratification among Lynch syndrome patients.

Cancer-specific expression quantitative loci are affected by expression dysregulation

Expression quantitative trait loci (eQTLs) have been touted as the missing piece that can bridge the gap between genetic variants and phenotypes. Over the past decade, we have witnessed a sharp rise of effort in the identification and application of eQTLs. The successful application of eQTLs relies heavily on their reproducibility. The current eQTL databases such as Genotype-Tissue Expression (GTEx) were populated primarily with eQTLs deriving from germline single nucleotide polymorphisms and normal tissue gene expression. The novel scenarios that employ eQTL models for prediction purposes often involve disease phenotypes characterized by altered gene expressions. To evaluate eQTL reproducibility across diverse data sources and the effect of disease-specific gene expression alteration on eQTL identification, we conducted an eQTL study using 5178 samples from The Cancer Genome Atlas (TCGA). We found that the reproducibility of eQTLs between normal and tumor tissues was low in terms of the number of shared eQTLs. However, among the shared eQTLs, the effect directions were generally concordant. This suggests that the source of the gene expression (normal or tumor tissue) has a strong effect on the detectable eQTLs and the effect direction of the eQTLs. Additional analyses demonstrated good directional concordance of eQTLs between GTEx and TCGA. Furthermore, we found that multi-tissue eQTLs may exert opposite effects across multiple tissue types. In summary, our results suggest that eQTL prediction models need to carefully address tissue and disease dependency of eQTLs. Tissue-disease-specific eQTL databases can afford more accurate prediction models for future studies.

Genetics of Obesity Traits: A Bivariate Genome-Wide Association Analysis

Previous genome-wide association studies on anthropometric measurements have identified more than 100 related loci, but only a small portion of heritability in obesity was explained. Here we present a bivariate twin study to look for the genetic variants associated with body mass index and waist-hip ratio, and to explore the obesity-related pathways in Northern Han Chinese. Cholesky decomposition model for 242 monozygotic and 140 dizygotic twin pairs indicated a moderate genetic correlation (r = 0.53, 95%CI: 0.42-0.64) between body mass index and waist-hip ratio. Bivariate genome-wide association analysis in 139 dizygotic twin pairs identified 26 associated SNPs with p < 10^-5. Further gene-based analysis found 291 nominally associated genes (P < 0.05), including F12, HCRTR1, PHOSPHO1, DOCK2, DOCK6, DGKB, GLP1R, TRHR, MMP1, GPR55, CCK, and OR2AK2, as well as 6 enriched gene-sets with FDR < 0.05. Expression quantitative trait loci analysis identified rs2242044 as a significant cis-eQTL in both the normal adipose-subcutaneous (P = 1.7 × 10^-9) and adipose-visceral (P = 4.4 × 10^-15) tissue. These findings may provide an important entry point to unravel genetic pleiotropy in obesity traits.

Pro-inflammatory fatty acid profile and colorectal cancer risk: A Mendelian randomisation analysis

BACKGROUND

While dietary fat has been established as a risk factor for colorectal cancer (CRC), associations between fatty acids (FAs) and CRC have been inconsistent. Using Mendelian randomisation (MR), we sought to evaluate associations between polyunsaturated (PUFA), monounsaturated (MUFA) and saturated FAs (SFAs) and CRC risk.

METHODS

We analysed genotype data on 9254 CRC cases and 18,386 controls of European ancestry. Externally weighted polygenic risk scores were generated and used to evaluate associations with CRC per one standard deviation increase in genetically defined plasma FA levels.

RESULTS

Risk reduction was observed for oleic and palmitoleic MUFAs (OROA = 0.77, 95% CI: 0.65-0.92, P = 3.9 × 10-3; ORPOA = 0.36, 95% CI: 0.15-0.84, P = 0.018). PUFAs linoleic and arachidonic acid had negative and positive associations with CRC respectively (ORLA = 0.95, 95% CI: 0.93-0.98, P = 3.7 × 10-4; ORAA = 1.05, 95% CI: 1.02-1.07, P = 1.7 × 10-4). The SFA stearic acid was associated with increased CRC risk (ORSA = 1.17, 95% CI: 1.01-1.35, P = 0.041).

CONCLUSION

Results from our analysis are broadly consistent with a pro-inflammatory FA profile having a detrimental effect in terms of CRC risk.