Search

Summary of 44 retained introns

Differential expression of 44 significant retained introns between races and four subtypes

BACKGROUND: Systemic inflammation is associated with survival outcomes in colon cancer. However, it is not well-known which systemic inflammatory marker is a powerful prognostic marker in patients with colon cancer. METHODS: A total of 4535 colon cancer patients were included in this study. We developed a novel prognostic index using a robust combination of seven systemic inflammation-associated blood features of the discovery set. The predictability and generality of the novel prognostic index were evaluated in the discovery, validation and replication sets. RESULTS: Among all combinations, the combination of albumin and monocyte count was the best candidate expression. The final formula of the proposed novel index is named the Prognostic Immune and Nutritional Index (PINI). The concordance index of PINI for overall and progression-free survival was the highest in the discovery, validation and replication sets compared to existing prognostic inflammatory markers. PINI was found to be a significant independent prognostic factor for both overall and progression-free survival. CONCLUSIONS: PINI is a novel prognostic index that has improved discriminatory power in colon cancer patients and appears to be superior to existing prognostic inflammatory markers. PINI can be utilised for decision-making regarding personalised treatment as the complement of the TNM staging system.

Background Various epigenetic factors are responsible for the non-genetic regulation on gene expression. The epigenetically dysregulated oncogenes or tumor suppressors by miRNA and/or DNA methylation are often observed in cancer cells. Each of these epigenetic regulators has been studied well in cancer progressions; however, their mutual regulatory relationship in cancer still remains unclear. In this study, we propose an integrative framework to systematically investigate epigenetic interactions between miRNA and methylation at the alternatively spliced mRNA level in bladder cancer. Each of these epigenetic regulators has been studied well in cancer progressions; however, their mutual regulatory relationship in cancer still remains unclear. Results The integrative analyses yielded 136 significant combinations (methylation, miRNA and isoform). Further, overall survival analysis on the 136 combinations based on methylation and miRNA, high and low expression groups resulted in 13 combinations associated with survival. Additionally, different interaction patterns were examined. Conclusions Our study provides a higher resolution of molecular insight into the crosstalk between two epigenetic factors, DNA methylation and miRNA. Given the importance of epigenetic interactions and alternative splicing in cancer, it is timely to identify and understand the underlying mechanisms based on epigenetic markers and their interactions in cancer, leading to alternative splicing with primary functional impact.

Background: Rare variants play an essential role in the etiology of cancer. In this study, we aim to characterize rare germline variants that impact the risk of cancer. Methods: We performed a genome-wide rare variant analysis using germline whole exome sequencing (WES) data derived from the Geisinger MyCode initiative to discover cancer predisposition variants. The case–control association analysis was conducted by binning variants in 5,538 patients with cancer and 7,286 matched controls in a discovery set and 1,991 patients with cancer and 2,504 matched controls in a validation set across nine cancer types. Further, The Cancer Genome Atlas (TCGA) germline data were used to replicate the findings. Results: We identified 133 significant pathway–cancer pairs (85 replicated) and 90 significant gene–cancer pairs (12 replicated). In addition, we identified 18 genes and 3 pathways that were associated with survival outcome across cancers (Bonferroni P < 0.05). Conclusions: In this study, we identified potential predisposition genes and pathways based on rare variants in nine cancers. Impact: This work adds to the knowledge base and progress being made in precision medicine.

STUDY DESIGN A case-control genome-wide association study (GWAS) on spondylosis. OBJECTIVE Leveraging Geisinger's MyCode initiative's multimodal dataset, we aimed to identify genetic associations with degenerative spine disease. SUMMARY OF BACKGROUND DATA Degenerative spine conditions are a leading cause of global disability; however, the genetic underpinnings of these conditions remain under-investigated. Previous studies using candidate-gene approach suggest a genetic risk for degenerative spine conditions, but large-scale GWASs are lacking. METHODS We identified 4434 patients with a diagnosis of spondylosis using ICD diagnosis codes with genotype data available. We identified a population-based control of 12,522 patients who did not have any diagnosis for osteoarthritis. A linear-mix, additive genetic model was employed to perform the genetic association tests adjusting for age, sex, and genetic principal components to account for the population structure and relatedness. Gene-based association tests were performed and heritability and genetic correlations with other traits were investigated. RESULTS We identified a genome-wide significant locus at rs12190551 (odds ratio = 1.034, 95% confidence interval 1.022-1.046, P = 8.5 × 10-9, minor allele frequency = 36.9%) located in the intron of BMP6. Additionally, NIPAL1 and CNGA1 achieved Bonferroni significance in the gene-based association tests. The estimated heritability was 7.19%. Furthermore, significant genetic correlations with pain, depression, lumbar spine bone mineral density, and osteoarthritis were identified. CONCLUSION We demonstrated the use of a massive database of genotypes combined with electronic health record data to identify a novel and significant association spondylosis. We also identified significant genetic correlations with pain, depression, bone mineral density, and osteoarthritis, suggesting shared genetic etiology and molecular pathways with these phenotypes.Level of Evidence: N/A.

Purpose The aim of this study was to compare the characteristic expression patterns of advanced periodontitis in 2 cohort data sets analyzed using different microarray platforms, and to identify differentially expressed genes (DEGs) through a meta-analysis of both data sets. Methods Twenty-two patients for cohort 1 and 40 patients for cohort 2 were recruited with the same inclusion criteria. The 2 cohort groups were analyzed using different platforms: Illumina and Agilent. A meta-analysis was performed to increase reliability by removing statistical differences between platforms. An integrative meta-analysis based on an empirical Bayesian methodology (ComBat) was conducted. DEGs for the integrated data sets were identified using the limma package to adjust for age, sex, and platform and compared with the results for cohorts 1 and 2. Clustering and pathway analyses were also performed. Results This study detected 557 and 246 DEGs in cohorts 1 and 2, respectively, with 146 and 42 significantly enriched gene ontology (GO) terms. Overlapping between cohorts 1 and 2 was present in 59 DEGs and 18 GO terms. However, only 6 genes from the top 30 enriched DEGs overlapped, and there were no overlapping GO terms in the top 30 enriched pathways. The integrative meta-analysis detected 34 DEGs, of which 10 overlapped in all the integrated data sets of cohorts 1 and 2. Conclusions The characteristic expression pattern differed between periodontitis and the healthy periodontium, but the consistency between the data sets from different cohorts and metadata was too low to suggest specific biomarkers for identifying periodontitis., Graphical Abstract

BackgroundPrevious studies primarily targeted the ability of polygenic risk scores (PRSs) to predict a specific disease, and only a few studies have investigated the association between genetic risk scores and cardiovascular (CV) mortality. We assessed PRSs for coronary artery disease (CAD) and type 2 diabetes (T2DM) as predictive factors for CV mortality, independent of traditional risk factors, and further investigated the additive effect between lifestyle behavior and PRS on CV mortality. MethodsWe used genetic and phenotypic data from UK Biobank participants aged 40–69 years at baseline, collected with standardized procedures. Genome-wide PRSs were constructed using >6 million genetic variants. Cox proportional hazard models were used to analyze the relationship between PRS and CV mortality with stratification by age, sex, disease status, and lifestyle behavior. ResultsOf 377,909 UK Biobank participants having European ancestry, 3,210 (0.8%) died due to CV disease during a median follow-up of 8.9 years. CV mortality risk was significantly associated with CAD PRS (low vs. very high genetic risk groups, CAD PRS hazard ratio [HR] 2.61 [2.02–3.36]) and T2DM PRS (HR 2.08 [1.58–2.73]), respectively. These relationships remained significant even after adjustment for a comprehensive range of demographic and clinical factors. In the very high genetic risk group, adherence to an unfavorable lifestyle was further associated with a substantially increased risk of CV mortality (favorable versus unfavorable lifestyle with very high genetic risk for CAD PRS, HR 8.31 [5.12–13.49]; T2DM PRS, HR 5.84 [3.39–10.04]). Across all genetic risk groups, 32.1% of CV mortality was attributable to lifestyle behavior (population attributable fraction [PAF] 32.1% [95% CI 28.8–35.3%]) and 14.1% was attributable to smoking (PAF 14.1% [95% CI 12.4–15.7%]). There was no evidence of significant interaction between PRSs and age, sex, or lifestyle behavior in predicting the risk of CV mortality. ConclusionsPRSs for CAD or T2DM and lifestyle behaviors are independent predictive factors for future CV mortality in the white, middle-aged population. PRS-based risk assessment could be useful to identify individuals who need intensive behavioral or therapeutic interventions to reduce the risk of CV mortality.

BackgroundDisease complications, the onset of secondary phenotypes given a primary condition, can exacerbate the long-term severity of outcomes. However, the exact cause of many of these cross-phenotype associations is still unknown. One potential reason is shared genetic etiology—common genetic drivers may lead to the onset of multiple phenotypes. Disease-disease networks (DDNs), where nodes represent diseases and edges represent associations between diseases, can provide an intuitive way of understanding the relationships between phenotypes. Using summary statistics from a phenome-wide association study (PheWAS), we can generate a corresponding DDN where edges represent shared genetic variants between diseases. Such a network can help us analyze genetic associations across the diseasome, the landscape of all human diseases, and identify potential genetic influences for disease complications.ResultsTo improve the ease of network-based analysis of shared genetic components across phenotypes, we developed the humaN disEase phenoType MAp GEnerator (NETMAGE), a web-based tool that produces interactive DDN visualizations from PheWAS summary statistics. Users can search the map by various attributes and select nodes to view related phenotypes, associated variants, and various network statistics. As a test case, we used NETMAGE to construct a network from UK BioBank (UKBB) PheWAS summary statistic data. Our map correctly displayed previously identified disease comorbidities from the UKBB and identified concentrations of hub diseases in the endocrine/metabolic and circulatory disease categories. By examining the associations between phenotypes in our map, we can identify potential genetic explanations for the relationships between diseases and better understand the underlying architecture of the human diseasome. Our tool thus provides researchers with a means to identify prospective genetic targets for drug design, using network medicine to contribute to the exploration of personalized medicine.

Background. The imputation of missingness is a key step in Electronic Health Records (EHR) mining, as it can significantly affect the conclusions derived from the downstream analysis in translational medicine. The missingness of laboratory values in EHR is not at random, yet imputation techniques tend to disregard this key distinction. Consequently, the development of an adaptive imputation strategy designed specifically for EHR is an important step in improving the data imbalance and enhancing the predictive power of modeling tools for healthcare applications. Method. We analyzed the laboratory measures derived from Geisinger&rsquo, s EHR on patients in three distinct cohorts&mdash, patients tested for Clostridioides difficile (Cdiff) infection, patients with a diagnosis of inflammatory bowel disease (IBD), and patients with a diagnosis of hip or knee osteoarthritis (OA). We extracted Logical Observation Identifiers Names and Codes (LOINC) from which we excluded those with 75% or more missingness. The comorbidities, primary or secondary diagnosis, as well as active problem lists, were also extracted. The adaptive imputation strategy was designed based on a hybrid approach. The comorbidity patterns of patients were transformed into latent patterns and then clustered. Imputation was performed on a cluster of patients for each cohort independently to show the generalizability of the method. The results were compared with imputation applied to the complete dataset without incorporating the information from comorbidity patterns. Results. We analyzed a total of 67,445 patients (11,230 IBD patients, 10,000 OA patients, and 46,215 patients tested for C. difficile infection). We extracted 495 LOINC and 11,230 diagnosis codes for the IBD cohort, 8160 diagnosis codes for the Cdiff cohort, and 2042 diagnosis codes for the OA cohort based on the primary/secondary diagnosis and active problem list in the EHR. Overall, the most improvement from this strategy was observed when the laboratory measures had a higher level of missingness. The best root mean square error (RMSE) difference for each dataset was recorded as &minus, 35.5 for the Cdiff, &minus, 8.3 for the IBD, and &minus, 11.3 for the OA dataset. Conclusions. An adaptive imputation strategy designed specifically for EHR that uses complementary information from the clinical profile of the patient can be used to improve the imputation of missing laboratory values, especially when laboratory codes with high levels of missingness are included in the analysis.

Background: Recent studies have found that women with obstetric disorders are at increased risk for a variety of long-term complications. However, the underlying pathophysiology of these connections remains undetermined. A network-based view incorporating knowledge of other diseases and genetic associations will aid our understanding of the role of genetics in pregnancy-related disease complications. Methods: We built a disease–disease network (DDN) using UK Biobank (UKBB) summary data from a phenome-wide association study (PheWAS) to elaborate multiple disease associations. We also constructed egocentric DDNs, where each network focuses on a pregnancy-related disorder and its neighboring diseases. We then applied graph-based semi-supervised learning (GSSL) to translate the connections in the egocentric DDNs to pathologic knowledge. Results: A total of 26 egocentric DDNs were constructed for each pregnancy-related phenotype in the UKBB. Applying GSSL to each DDN, we obtained complication risk scores for additional phenotypes given the pregnancy-related disease of interest. Predictions were validated using co-occurrences derived from UKBB electronic health records. Our proposed method achieved an increase in average area under the receiver operating characteristic curve (AUC) by a factor of 1.35 from 55.0% to 74.4% compared to the use of the full DDN. Conclusion: Egocentric DDNs hold promise as a clinical tool for the network-based identification of potential disease complications for a variety of phenotypes.

Background: We aimed to investigate the associations between polygenic risk score (PRS) for T2DM and future risk of cardiovascular disease (CVD) in middle-aged UK Biobank participants of European ancestry. Methods: We used data from UK Biobank participants aged 40-69 years at baseline, registered from 22 assessment centers across the United Kingdom with standardized procedures. Genome-wide PRSs were constructed using >9 million genetic variants, and newly developed myocardial infarction (MI), ischemic stroke, and cardiovascular (CV) mortality were considered as outcomes. Cox proportional hazard models were used to analyze the relationship between PRSs and CV mortality with stratification by age, sex, and lifestyle behavior. Results: Of 362,405 participants having European ancestry, 8,787 (2.4%) experienced cardiovascular events and 2,414 (0.7%) died due to CVD during a median follow-up of 8.9 years. Risk of CV mortality and of myocardial infarction (MI) were both significantly associated with T2DM PRS (low vs. very high T2DM genetic risk groups, hazard ratio [HR] of CV mortality 1.73 [1.23-2.44] and HR of MI 1.55 [1.24-1.96], P Conclusion: PRS for T2DM is an independent predictive factor for future CVD and associated mortality in the white, middle-aged population. PRS-based genetic risk assessment could be useful to identify individuals who need intensive behavioral or therapeutic interventions to reduce the risk of CVD. Disclosure S. Jung: None. D. Kim: None. M. Shivakumar: None. H. Won: None. J. Yun: None.

OBJECTIVETo assess the prognostic ability of polygenic risk scores (PRSs) for coronary artery disease (CAD) and type 2 diabetes mellitus (T2DM) for cardiovascular (CV) mortality, independent of traditional risk factors, and further investigate the additive effect between lifestyle behavior and PRS on CV mortality.DESIGNProspective population-based cohort study.SETTINGUK Biobank.PARTICIPANTSA total 377,909 unrelated participants of white British descent were included in the analyses from the UK Biobank cohort.MAIN OUTCOME MEASURESGenome-wide PRSs were constructed using >6 million genetic variants. We stratified patients into four PRS risk groups: low (0 to 19th percentile), intermediate (20 to 79th percentile), high (80 to 98th percentile), and very high (99th percentile). We defined a favorable and unfavorable lifestyle with four modifiable lifestyle components, including smoking, obesity, physical activity, and diet. Cox proportional hazard models were used to analyze the relationship between PRS and CV mortality with stratification by age, sex, disease status, and lifestyle behavior.RESULTSOf 377,909 UK Biobank participants having European ancestry, 3,210 (0.8%) died due to CV disease during a median follow-up of 8.9 years. CV mortality risk was significantly associated with CAD PRS (low vs. very high genetic risk groups, CAD PRS hazard ratio [HR] 2.61 [2.02 to 3.36]) and T2DM PRS (HR 2.08 [1.58 to 2.73]), respectively. These relationships remained significant even after adjustment for a comprehensive range of demographic and clinical factors. In the very high genetic risk group, adherence to an unfavorable lifestyle was further associated with a substantially increased risk of CV mortality (favorable versus unfavorable lifestyle with very high genetic risk for CAD PRS, HR 8.31 [5.12 to 13.49]; T2DM PRS, HR 5.84 [3.39 to 10.04]). Across all genetic risk groups, 32.1% of CV mortality was attributable to lifestyle behavior (population attributable fraction [PAF] 32.1% [95% CI 28.8 to 35.3%]) and 14.1% was attributable to smoking (PAF 14.1% [95% CI 12.4 to 15.7%]). There was no evidence of significant interaction between PRSs and age, sex, or lifestyle behavior in predicting the risk of CV mortality.CONCLUSIONPRSs for CAD or T2DM and lifestyle behaviors are independent predictive factors for future CV mortality in the white, middle-aged population. PRS-based risk assessment could be useful to identify individuals who need intensive behavioral or therapeutic interventions to reduce the risk of CV mortality.Summary BoxWhat is already known on this topicPolygenic risk scores quantify the inherited risk conferred by the cumulative impact of common variants into a quantitative risk estimate.Previous studies primarily targeted the ability of polygenic risk scores to predict a specific disease, and only a few studies have investigated the association between genetic risk scores and cardiovascular mortality.The majority of previous analyses calculated polygenic risk scores from only a small number of genetic variants or adjusted for only a few risk factors, and no studies have examined whether the association of polygenic risk score with cardiovascular mortality differs by lifestyle behavior.What this study addsGenetic risk and lifestyle are independent predictive factors for cardiovascular mortality, even after adjustment for a comprehensive range of demographic and clinical factors.A healthy lifestyle is associated with relative risk reduction for cardiovascular mortality across all genetic risk categories, a finding that indicates the potential benefit of intensive lifestyle modification in overcoming genetic risk for cardiovascular mortality.

Background inflammatory status indicators have been reported as prognostic biomarkers of colorectal cancer (CRC). However, since inflammatory interactions with the colon involve various modes of action, the biological mechanism linking inflammation and CRC prognosis has not been fully elucidated. We comprehensively evaluated the predictive roles of the expression and methylation levels of inflammation-related genes for CRC prognosis and their pathophysiological associations. Method. An integrative analysis of 247 patients with stage I-III CRC from The Cancer Genome Atlas was conducted. Lasso-penalized Cox proportional hazards regression (Lasso-Cox) and statistical Cox proportional hazard regression (CPH) were used for the analysis. Results. Models to predict overall survival were designed with respective combinations of clinical variables, including age, sex, stage, gene expression, and methylation. An integrative model combining expression, methylation, and clinical features performed better (median C-index = 0.756) than the model with clinical features alone (median C-index = 0.726). Based on multivariate CPH with features from the best model, the methylation levels of CEP250, RAB21, and TNPO3 were significantly associated with overall survival. They did not share any biological process in functional networks. The 5-year survival rate was 29.8% in the low methylation group of CEP250 and 79.1% in the high methylation group (p &lt, 0.001). Conclusion. Our study results implicate the importance of integrating expression and methylation information along with clinical information in the prediction of survival. CEP250, RAB21, and TNPO3 in the prediction model might have a crucial role in CRC prognosis and further improve our understanding of potential mechanisms linking inflammatory reactions and CRC progression.

Simple Summary The objectives of this study were to identify rare germline variant associations in urothelial carcinoma of the bladder (UC) incidence and to determine its association with clinical outcomes. Our analysis reveals that individuals with MR1 rare germline variants had significantly worse OS than those without any, and those with ADGRL2 variants were slightly more likely to have UC compared to a control cohort matched for age, sex, and smoking status. These associations, using models incorporating known environmental covariates and using well-defined clinical phenotypes, highlight several candidates for prognostic indicator genes for the differential presentation of UC. These associations highlight several candidate genes that have the potential to explain clinical disparities in UC and predict UC outcomes. Abstract Urothelial carcinoma of the bladder (UC) is the fifth most common cancer in the United States. Germline variants, especially rare germline variants, may account for a portion of the disparity seen among patients in terms of UC incidence, presentation, and outcomes. The objectives of this study were to identify rare germline variant associations in UC incidence and to determine its association with clinical outcomes. Using exome sequencing data from the DiscovEHR UC cohort (n = 446), a European-ancestry, North American population, the complex influence of germline variants on known clinical phenotypes were analyzed using dispersion and burden metrics with regression tests. Outcomes measured were derived from the electronic health record (EHR) and included UC incidence, age at diagnosis, and overall survival (OS). Consequently, key rare variant association genes were implicated in MR1 and ADGRL2. The Kaplan–Meier survival analysis reveals that individuals with MR1 germline variants had significantly worse OS than those without any (log-rank p-value = 3.46 × 10−7). Those with ADGRL2 variants were found to be slightly more likely to have UC compared to a matched control cohort (FDR q-value = 0.116). These associations highlight several candidate genes that have the potential to explain clinical disparities in UC and predict UC outcomes.

Summary Osteoarthritis affects over 300 million people worldwide. Here, we conduct a genome-wide association study meta-analysis across 826,690 individuals (177,517 with osteoarthritis) and identify 100 independently associated risk variants across 11 osteoarthritis phenotypes, 52 of which have not been associated with the disease before. We report thumb and spine osteoarthritis risk variants and identify differences in genetic effects between weight-bearing and non-weight-bearing joints. We identify sex-specific and early age-at-onset osteoarthritis risk loci. We integrate functional genomics data from primary patient tissues (including articular cartilage, subchondral bone, and osteophytic cartilage) and identify high-confidence effector genes. We provide evidence for genetic correlation with phenotypes related to pain, the main disease symptom, and identify likely causal genes linked to neuronal processes. Our results provide insights into key molecular players in disease processes and highlight attractive drug targets to accelerate translation., Graphical abstract, Highlights • A multicohort study identifies 52 previously unknown osteoarthritis genetic risk variants • Similarities and differences in osteoarthritis genetic risk depend on joint sites • Osteoarthritis genetic components are associated with pain-related phenotypes • High-confidence effector genes highlight potential targets for drug intervention, A multicohort genome-wide association meta-analysis of osteoarthritis highlights the impact of joint site types on the features of genetic risk variants and the link between osteoarthritis genetics and pain-related phenotypes, pointing toward potential targets for therapeutic intervention.

SummaryGiven genetic associations from a PheWAS, a disease-disease network can be constructed where nodes represent phenotypes and edges represent shared genetic associations between phenotypes. To improve the accessibility of the visualization of shared genetic components across phenotypes, we developed the humaN-disEase phenoType MAp GEnerator (NETMAGE), a web-based tool that produces interactive phenotype network visualizations from summarized PheWAS results. Users can search the map by a variety of attributes, and they can select nodes to view information such as related phenotypes, associated SNPs, and other network statistics. As a test case, we constructed a network using UK BioBank PheWAS summary data. By examining the associations between phenotypes in our map, we can potentially identify novel instances of pleiotropy, where loci influence multiple phenotypic traits. Thus, our tool provides researchers with a means to identify prospective genetic targets for drug design, contributing to the exploration of personalized medicine.Availability and implementationOur service runs athttps://hdpm.biomedinfolab.com. Source code can be downloaded athttps://github.com/dokyoonkimlab/netmage.Contactdokyoon.kim@pennmedicine.upenn.eduSupplementary informationSupplementary data and user guide are available atBioinformaticsonline.

Motivation To better understand the molecular features of cancers, a comprehensive analysis using multi-omics data has been conducted. In addition, a pathway activity inference method has been developed to facilitate the integrative effects of multiple genes. In this respect, we have recently proposed a novel integrative pathway activity inference approach, iDRW and demonstrated the effectiveness of the method with respect to dichotomizing two survival groups. However, there were several limitations, such as a lack of generality. In this study, we designed a directed gene–gene graph using pathway information by assigning interactions between genes in multiple layers of networks. Results As a proof-of-concept study, it was evaluated using three genomic profiles of urologic cancer patients. The proposed integrative approach achieved improved outcome prediction performances compared with a single genomic profile alone and other existing pathway activity inference methods. The integrative approach also identified common/cancer-specific candidate driver pathways as predictive prognostic features in urologic cancers. Furthermore, it provides better biological insights into the prioritized pathways and genes in an integrated view using a multi-layered gene–gene network. Our framework is not specifically designed for urologic cancers and can be generally applicable for various datasets. Availability and implementation iDRW is implemented as the R software package. The source codes are available at https://github.com/sykim122/iDRW. Supplementary information Supplementary data are available at Bioinformatics online.

sBackgroundThe expanding use of the phenome-wide association study (PheWAS) faces challenges in the context of using International Classification of Diseases billing codes for phenotype definition, imbalanced study population ethnicity, and constrained application of the results to clinical practice or research.MethodsWe performed a PheWAS utilizing deep phenotypes corroborated by comprehensive health check-ups in a Korean population, along with trans-ethnic comparisons through the UK Biobank and Biobank Japan Project. Network analysis, visualization of cross-phenotype mapping, and causal inference mapping with Mendelian randomization were conducted in order to make robust, clinically applicable interpretations.ResultsOf the 136 phenotypes extracted from the health check-up database, the PheWAS associated 65 phenotypes with 14,101 significant variants (P < 4.92×10−10). In the association study for body mass index, our population showed 583 exclusive loci relative to the Japanese population and 669 exclusive loci relative to the European population. In the meta-analysis with Korean and Japanese populations, 72.5% of phenotypes had uniquely significant variants. Tumor markers and hematologic phenotypes had a high degree of phenotype-phenotype pairs. By Mendelian randomization, one skeletal muscle mass phenotype was causal and two were outcomes. Among phenotype pairs from the genotype-driven cross-phenotype associations, 71.65% also demonstrated penetrance in correlation analysis using a clinical database.ConclusionsThis comprehensive analysis of PheWAS results based on a health check-up database will provide researchers and clinicians with a panoramic overview of the networks among multiple phenotypes and genetic variants, laying groundwork for the practical application of precision medicine.

Background Inflammatory status indicators have been reported as a prognostic biomarker of colorectal cancer (CRC). However, since the inflammatory interactions with colon involve various modes of action, the biological mechanism to link inflammation and CRC prognosis is not fully elucidated. We comprehensively evaluated the predictive role of the expression and methylation level of inflammation-related genes for CRC prognosis and their pathophysiological associations. Method An integrative analysis was conducted on 247 patients of stage I-III CRC from The Cancer Genome Atlas. Lasso-penalized Cox proportional hazards regression (Lasso-Cox) and statistical Cox proportional hazard regression (CPH) were used for analysis. Result Models to predict overall survival were designed with respective combinations of clinical variables, including age, sex, stage, gene expression, and methylation. An integrative model combining expression, methylation, and clinical features had the highest performance (median C-index=0.756), compared to the model with clinical features alone (median C-index=0.726). By multivariate CPH with features from the best model, methylation levels of CEP250, RAB21 and TNPO3 were significantly associated with overall survival. They did not share any biological process in functional networks. The 5-year survival rate was 29.8% in a low methylation group of CEP250 and 79.1% in a high (P

As part of a data mining competition, a training and test set of laboratory test data about patients with and without surgical site infection (SSI) were provided. The task was to develop predictive models with training set and identify patients with SSI in the no label test set. Lab test results are vital resources that guide healthcare providers make decisions about all aspects of surgical patient management. Many machine learning models were developed after pre-processing and imputing the lab tests data and only the top performing methods are discussed. Overall, RANDOM FOREST algorithms performed better than Support Vector Machine and Logistic Regression. Using a set of 74 lab tests, with RF, there were only 4 false positives in the training set and predicted 35 out of 50 SSI patients in the test set (Accuracy 0.86, Sensitivity 0.68, and Specificity 0.91). Optimal ways to address healthcare data quality concerns and imputation methods as well as newer generalizable algorithms need to be explored further to decipher new associations and knowledge among laboratory biomarkers and SSI.

Regulation of gene expression by DNA methylation in gene promoter regions is well studied; however, the effects of methylation in the gene body (exons and introns) on gene expression are comparatively understudied. Recently, hypermethylation has been implicated in the inclusion of alternatively spliced exons; moreover, exon recognition can be enhanced by recruiting the methyl-CpG-binding protein (MeCP2) to hypermethylated sites. This study examines whether the methylation status of an intron is correlated with how frequently the intron is retained during splicing using DNA methylation and RNA sequencing data from breast cancer tissue specimens in The Cancer Genome Atlas. Interestingly, hypomethylation of introns is correlated with higher levels of intron expression in mRNA and the methylation level of an intron is inversely correlated with its retention in mRNA from the gene in which it is located. Furthermore, significant population differences were observed in the methylation level of retained introns. In African-American donors, retained introns were not only less methylated compared to European-American donors, but also were more highly expressed. This underscores the need for understanding epigenetic differences in populations and their correlation with breast cancer is an important step toward achieving personalized cancer care. Implications: This research contributes to the understanding of how epigenetic markers in the gene body communicate with the transcriptional machinery to control transcript diversity and differential biological response to changes in methylation status could underlie some of the known, yet unexplained, disparities in certain breast cancer patient populations. Mol Cancer Res; 16(3); 461–9. ©2018 AACR.

As part of a data mining competition, a training and test set of laboratory test data about patients with and without surgical site infection (SSI) were provided. The task was to develop predictive models with training set and identify patients with SSI in the no label test set. Lab test results are vital resources that guide healthcare providers make decisions about all aspects of surgical patient management. Many machine learning models were developed after pre-processing and imputing the lab tests data and only the top performing methods are discussed. Overall, RANDOM FOREST algorithms performed better than Support Vector Machine and Logistic Regression. Using a set of 74 lab tests, with RF, there were only 4 false positives in the training set and predicted 35 out of 50 SSI patients in the test set (Accuracy 0.86, Sensitivity 0.68, and Specificity 0.91). Optimal ways to address healthcare data quality concerns and imputation methods as well as newer generalizable algorithms need to be explored further to decipher new associations and knowledge among laboratory biomarkers and SSI.

Background: Gestational diabetes (GDM) has profound effects on the intrauterine metabolic milieu, induces marked abnormalities in fetal glucose and insulin secretion and is linked to obesity and diabetes in the offspring, but the mechanisms remain largely unknown. Epigenetic modifications in stems cells may be one mechanism by which an in utero exposure can lead to the development of diabetes and obesity later in life. Objective: To identify novel pathways contributing to the development of diabetes and obesity in offspring exposed to GDM in utero by integrating data generated from transcriptome and methylome analysis from second trimester human amniocytes exposed to GDM in utero. Methods: We analyzed RNAseq and genome wide DNA methylation data (ERRBS) generated from second trimester amniocytes obtained from women with GDM (n=14). Amniocytes have stem cells-like characteristics and are derived from the fetus. Expression data of 22,271 genes were retrieved from RNAseq data. CpGs with significant changes in DNA methylation were mapped into 20,028 genes by collapsing methylation probes into promoter and gene regions. To better understand the associations among diverse gene sets or among gene sets and GDM,we first constructed two weighted co-expression networks from RNAseq and DNA methylation data, respectively. Then, two co-expression networks were integrated using a linear combination. With the integrated co-expression network, graph-based label propagation algorithm was employed to prioritize GDM-associated genes. Results: From the differential expression analysis between GDM and control, the top 20 query genes, including 11 genes and 9 methylated genes, were selected for label propagation. Finally, the top 100 genes were picked up for the pathway enrichment testusing an over-representation analysis approach. Significantly enriched pathways included: Interferon Signaling, N-glycan Antennae Elongation, Sphingolipid Pathway and Metabolism, Classical Complement Pathway, Complement and Coagulation Cascades, Tryptophan Metabolism, Peroxisomal Protein Import, Unsaturated Fatty Acid Metabolism, Complement Activation, Human Innate Immune Response, Ceramide Metabolism, Fertilization Pathway, Keratan Sulfate Biosynthesis Pathway, Transport to the Golgi and Modification Pathway (FDR q Conclusion: Using a novel bioinformatic approach to synthesize transcriptome and methylome data derived from human amniocytes exposed to GDM in utero, we identified potential pathways involved in programming of diabetes and obesity in offspring including pathways involving the immune response, complex lipid metabolism, the complement pathway, and protein transport and processing. Further investigation of these pathways may yield new mechanisms contributing to diabetes and obesity.

Rare variants play an essential role in the etiology of cancer and characterizing rare germline variants that impact the risk of cancer is an ongoing challenge. We performed a genome-wide rare variant analysis using germline whole exome sequencing (WES) data derived from the Geisinger MyCode initiative to discover cancer predisposition variants. The analysis was conducted using 5,538 cancer patients and 7,286 matched controls in a discovery set and 1,991 cancer patients and 2,504 matched controls in a validation set across nine cancer types. The variants in the WES data were annotated and filtered for pathogenic and likely pathogenic variants using ClinVar and Variant Effect Predictor (VEP). After filtering, case-control association tests were conducted on each cancer type separately by binning the pathogenic and likely pathogenic rare variants into genes and pathways using SKAT-O. We discovered 87 genes and 106 pathways significantly associated with cancer (Bonferroni-corrected P < 0.05) out of which seven genes and 26 pathways replicated from the validation set (suggestive threshold P < 0.05). Further, four genes and 21 pathways were discovered to be associated with multiple cancers (Bonferroni-corrected P < 0.05). Additionally, we identified 13 genes and two pathways associated with survival outcome across seven cancers (Bonferroni-corrected P < 0.05), where two genes, PCDHB8 and DCHS2, were also associated with survival outcome in TCGA data. In summary, we conducted one of the largest pan-cancer association studies using germline data derived from a single hospital system to find novel predisposition genes and pathways associated with nine cancers. Our results can inform future guidelines for germline genetic testing in cancer, which will be helpful in screening for cancer high-risk patients. This work adds to the knowledge base and progress being made in precision medicine.

Since prostate cancer is highly heritable, common variants associated with prostate cancer have been studied in various populations, including those in Korea. However, rare and low-frequency variants have a significant influence on the heritability of the disease. The contributions of rare variants to prostate cancer susceptibility have not yet been systematically evaluated in a Korean population. In this work, we present a large-scale exome-wide rare variant analysis of 7,258 individuals (985 cases with prostate cancer and 6,273 controls). In total, 19 rare variant loci spanning 7 genes contributed to an association with prostate cancer susceptibility. In addition to replicating previously known susceptibility genes (e.g., CDYL2, MST1R, GPER1, and PARD3B), 3 novel genes were identified (FDR q ENTPD3-AS1, LOC102724438, and protein-coding gene SPATA3. Additionally, 6 pathways were identified based on identified variants and genes, including estrogen signaling pathway, signaling by MST1, IL-15 production, MSP-RON signaling pathway, and IL-12 signaling and production in macrophages, which are known to be associated with prostate cancer. In summary, we report novel genes and rare variants that potentially play a role in prostate cancer susceptibility in the Korean population. These observations demonstrated a path towards one of the fundamental goals of precision medicine, which is to identify biomarkers for a subset of the population with a greater risk of disease than others.

Background: Carcinoembryonic antigen (CEA) is one of the most widely used tumor markers in colorectal cancer (CRC). High preoperative concentrations of CEA correlate with poor survival outcomes. Serial CEA measurements have been used for detecting recurrent CRC. However, the genomic characteristics of primary CRC according to the blood CEA level are still not clear. Though patients with low CEA levels are associated with good prognosis, some of these patients have progression of their tumor. It would aid in applying careful surveillance to identify a subset of patients at high risk for poor prognosis among those with low CEA level. In this study, the different types of molecular profiles were compared between CRC with low and high serum CEA levels. In addition, we compared the genomic characteristics between CRC without progression and with progression among patients with low CEA levels. Methods: Using data acquired from The Cancer Genome Atlas (TCGA), we performed survival and genomic analyses on 381 CRC patient samples with serum CEA levels. Differential mRNA expression, miRNA expression and DNA methylation were examined between CRC with low serum CEA levels (low CEA group; ≤ 5 ng/mL) and high serum CEA levels (high CEA group; > 5 ng/mL). Among the 140 patients with low CEA levels, the same analyses were performed between CRC without progression and with progression. Results: High CEA group was associated with higher pathologic stage, more venous and lymphatic invasion. High CEA group had worse overall and progression-free survival than low CEA group (p Conclusions: In this study, we have identified different molecular features between primary CRC with low and high CEA levels. Notably, specific genomic profiles may contribute to discriminating patients with poor outcomes in CRC with low CEA level. Citation Format: Ji Won Park, Manu Shivakumar, Eun Kyung Choe, Dokyoon Kim. Genomic characterization of primary tumor in colorectal cancer according to serum carcinoembryonic antigen level [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3626.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

BackgroundLarge-scale collaborative precision medicine initiatives (e.g., The Cancer Genome Atlas (TCGA)) are yielding rich multi-omics data. Integrative analyses of the resulting multi-omics data, such as somatic mutation, copy number alteration (CNA), DNA methylation, miRNA, gene expression, and protein expression, offer the tantalizing possibilities of realizing the potential of precision medicine in cancer prevention, diagnosis, and treatment by substantially improving our understanding of underlying mechanisms as well as the discovery of novel biomarkers for different types of cancers. However, such analyses present a number of challenges, including the heterogeneity of data types, and the extreme high-dimensionality of omics data.MethodsIn this study, we propose a novel framework for integrating multi-omics data based on multi-view feature selection, an emerging research problem in machine learning research. We also present a novel multi-view feature selection algorithm, MRMR-mv, which adapts the well-known Min-Redundancy and Maximum-Relevance (MRMR) single-view feature selection algorithm for the multi-view settings.ResultsWe report results of experiments on the task of building a predictive model of cancer survival from an ovarian cancer multi-omics dataset derived from the TCGA database. Our results suggest that multi-view models for predicting ovarian cancer survival outperform both view-specific models (i.e., models trained and tested using one multi-omics data source) and models based on two baseline data fusion methods.ConclusionsOur results demonstrate the potential of multi-view feature selection in integrative analyses and predictive modeling from multi-omics data.

Background Alzheimer’s disease (AD) is one of the most common neurodegenerative diseases that causes problems related to brain function. To some extent it is understood on a molecular level how AD arises, however there are a lack of biomarkers that can be used for early diagnosis. Two popular methods to identify AD-related biomarkers use genetics and neuroimaging. Genes and neuroimaging phenotypes have provided some insights as to the potential for AD biomarkers. While the field of imaging-genomics has identified genetic features associated with structural and functional neuroimaging phenotypes, it remains unclear how variants that affect splicing could be important for understanding the genetic etiology of AD. Methods In this study, rare variants (minor allele frequency

Imputation is a key step in Electronic Health Records-mining as it can significantly affect the conclusions derived from the downstream analysis. There are three main categories that explain the missingness in clinical settings–incompleteness, inconsistency, and inaccuracy–and these can capture a variety of situations: the patient did not seek treatment, the health care provider did not enter the information, etc. We used EHR data from patients diagnosed with Inflammatory Bowel Disease from Geisinger Health System to design a novel imputation that focuses on a complex phenotype. Our approach is based on latent-based analysis integrated with clustering to group patients based on their comorbidities before imputation. IBD is a chronic illness of unclear etiology and without a complete cure. We have taken advantage of the complexity of IBD to pre-process the EHR data of 10,498 IBD patients and show that imputation can be improved using shared latent comorbidities. The R code and sample simulated input data will be available at a future time.

Alzheimer's disease (AD) is a neurodegenerative disorder with few biomarkers even though it impacts a relatively large portion of the population and is predicted to affect significantly more individuals in the future. Neuroimaging has been used in concert with genetic information to improve our understanding in relation to how AD arises and how it can be potentially diagnosed. Additionally, evidence suggests synonymous variants can have a functional impact on gene regulatory mechanisms, including those related to AD. Some synonymous codons are preferred over others leading to a codon bias. The bias can arise with respect to codons that are more or less frequently used in the genome. A bias can also result from optimal and non-optimal codons, which have stronger and weaker codon anti-codon interactions, respectively. Although association tests have been utilized before to identify genes associated with AD, it remains unclear how codon bias plays a role and if it can improve rare variant analysis. In this work, rare variants from whole-genome sequencing from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort were binned into genes using BioBin. An association analysis of the genes with AD-related neuroimaging biomarker was performed using SKAT-O. While using all synonymous variants we did not identify any genomewide significant associations, using only synonymous variants that affected codon frequency we identified several genes as significantly associated with the imaging phenotype. Additionally, significant associations were found using only rare variants that contains an optimal codon in among minor alleles and a non-optimal codon in the major allele. These results suggest that codon bias may play a role in AD and that it can be used to improve detection power in rare variant association analysis.