Binze Li1, Julian Dong2, Jiaqi Yu3, Yuqi Fan4, Lulu Shang5, Xiang Zhou5,6, Yongsheng Bai7,8. 1. Bellaire High School, 5100 Maple St, Bellaire, TX, 77401, USA. 2. Northville High School, 45700 Six Mile Road, Northville, MI, 48168, USA. 3. College Preparatory School, 6100 Broadway, Oakland, CA, 94618, USA. 4. The Master's Academy, 1500 Lukas Ln, Oviedo, FL, 32765, USA. 5. Department of Biostatistics, University of Michigan, Ann Arbor, MI, 48109, USA. 6. Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, 48109, USA. 7. Department of Biology, Eastern Michigan University, Ypsilanti, MI, 48197, USA. yongshengbaicool@gmail.com. 8. Next-Gen Intelligent Science Training, Ann Arbor, MI, 48105, USA. yongshengbaicool@gmail.com.
Abstract
BACKGROUND: Understanding gene regulation is important but difficult. Elucidating tissue-specific gene regulation mechanism is even more challenging and requires gene co-expression network assembled from protein-protein interaction, transcription factor and gene binding, and post-transcriptional regulation (e.g., miRNA targeting) information. The miRNA binding affinity could therefore be changed by SNP(s) located at the 3' untranslated regions (3'UTR) of the target messenger RNA (mRNA) which miRNA(s) interacts with. Genome-wide association study (GWAS) has reported significant numbers of loci hosting SNPs associated with many traits. The goal of this study is to pinpoint GWAS functional variants located in 3'UTRs and elucidate if the genes harboring these variants along with their targeting miRNAs are associated with genetic traits relevant to certain tissues. METHODS: By applying MIGWAS, CoCoNet, ANNOVAR, and DAVID bioinformatics software and utilizing the gene expression database (e.g. GTEx data) to study GWAS summary statistics for 43 traits from 28 GWAS studies, we have identified a list of miRNAs and targeted genes harboring 3'UTR variants, which could contribute to trait-relevant tissue over miRNA-target gene network. RESULTS: Our result demonstrated that strong association between traits and tissues exists, and in particular, the Primary Biliary Cirrhosis (PBC) trait has the most significant p-value for all 180 tissues among all 43 traits used for this study. We reported SNPs located in 3'UTR regions of genes (SFMBT2, ZC3HAV1, and UGT3A1) targeted by miRNAs for PBC trait and its tissue association network. After employing Gene Ontology (GO) analysis for PBC trait, we have also identified a very important miRNA targeted gene over miRNA-target gene network, PFKL, which encodes the liver subunit of an enzyme. CONCLUSIONS: The non-coding variants identified from GWAS studies are casually assumed to be not critical to translated protein product. However, 3' untranslated regions (3'UTRs) of genes harbor variants can often change the binding affinity of targeting miRNAs playing important roles in protein translation degree. Our study has shown that GWAS variants could play important roles on miRNA-target gene networks by contributing the association between traits and tissues. Our analysis expands our knowledge on trait-relevant tissue network and paves way for future human disease studies.
BACKGROUND: Understanding gene regulation is important but difficult. Elucidating tissue-specific gene regulation mechanism is even more challenging and requires gene co-expression network assembled from protein-protein interaction, transcription factor and gene binding, and post-transcriptional regulation (e.g., miRNA targeting) information. The miRNA binding affinity could therefore be changed by SNP(s) located at the 3' untranslated regions (3'UTR) of the target messenger RNA (mRNA) which miRNA(s) interacts with. Genome-wide association study (GWAS) has reported significant numbers of loci hosting SNPs associated with many traits. The goal of this study is to pinpoint GWAS functional variants located in 3'UTRs and elucidate if the genes harboring these variants along with their targeting miRNAs are associated with genetic traits relevant to certain tissues. METHODS: By applying MIGWAS, CoCoNet, ANNOVAR, and DAVID bioinformatics software and utilizing the gene expression database (e.g. GTEx data) to study GWAS summary statistics for 43 traits from 28 GWAS studies, we have identified a list of miRNAs and targeted genes harboring 3'UTR variants, which could contribute to trait-relevant tissue over miRNA-target gene network. RESULTS: Our result demonstrated that strong association between traits and tissues exists, and in particular, the Primary Biliary Cirrhosis (PBC) trait has the most significant p-value for all 180 tissues among all 43 traits used for this study. We reported SNPs located in 3'UTR regions of genes (SFMBT2, ZC3HAV1, and UGT3A1) targeted by miRNAs for PBC trait and its tissue association network. After employing Gene Ontology (GO) analysis for PBC trait, we have also identified a very important miRNA targeted gene over miRNA-target gene network, PFKL, which encodes the liver subunit of an enzyme. CONCLUSIONS: The non-coding variants identified from GWAS studies are casually assumed to be not critical to translated protein product. However, 3' untranslated regions (3'UTRs) of genes harbor variants can often change the binding affinity of targeting miRNAs playing important roles in protein translation degree. Our study has shown that GWAS variants could play important roles on miRNA-target gene networks by contributing the association between traits and tissues. Our analysis expands our knowledge on trait-relevant tissue network and paves way for future human disease studies.
Authors: Hilary K Finucane; Brendan Bulik-Sullivan; Alexander Gusev; Gosia Trynka; Yakir Reshef; Po-Ru Loh; Verneri Anttila; Han Xu; Chongzhi Zang; Kyle Farh; Stephan Ripke; Felix R Day; Shaun Purcell; Eli Stahl; Sara Lindstrom; John R B Perry; Yukinori Okada; Soumya Raychaudhuri; Mark J Daly; Nick Patterson; Benjamin M Neale; Alkes L Price Journal: Nat Genet Date: 2015-09-28 Impact factor: 38.330
Authors: Joseph K Pickrell; Tomaz Berisa; Jimmy Z Liu; Laure Ségurel; Joyce Y Tung; David A Hinds Journal: Nat Genet Date: 2016-05-16 Impact factor: 38.330