The Genotype-Tissue Expression (GTEx) project was established to characterize genetic effects on the transcriptome across human tissues and to link these regulatory mechanisms to trait and disease associations. Here, we present analyses of the version 8 data, examining 15,201 RNA-sequencing samples from 49 tissues of 838 postmortem donors. We comprehensively characterize genetic associations for gene expression and splicing in cis and trans, showing that regulatory associations are found for almost all genes, and describe the underlying molecular mechanisms and their contribution to allelic heterogeneity and pleiotropy of complex traits. Leveraging the large diversity of tissues, we provide insights into the tissue specificity of genetic effects and show that cell type composition is a key factor in understanding gene regulatory mechanisms in human tissues.

Scalable, integrative methods to understand mechanisms that link genetic variants with phenotypes are needed. Here we derive a mathematical expression to compute PrediXcan (a gene mapping approach) results using summary data (S-PrediXcan) and show its accuracy and general robustness to misspecified reference sets. We apply this framework to 44 GTEx tissues and 100+ phenotypes from GWAS and meta-analysis studies, creating a growing public catalog of associations that seeks to capture the effects of gene expression variation on human phenotypes. Replication in an independent cohort is shown. Most of the associations are tissue specific, suggesting context specificity of the trait etiology. Colocalized significant associations in unexpected tissues underscore the need for an agnostic scanning of multiple contexts to improve our ability to detect causal regulatory mechanisms. Monogenic disease genes are enriched among significant associations for related traits, suggesting that smaller alterations of these genes may cause a spectrum of milder phenotypes.

The excision of introns from pre-mRNA is an essential step in mRNA processing. We developed LeafCutter to study sample and population variation in intron splicing. LeafCutter identifies variable splicing events from short-read RNA-seq data and finds events of high complexity. Our approach obviates the need for transcript annotations and circumvents the challenges in estimating relative isoform or exon usage in complex splicing events. LeafCutter can be used both for detecting differential splicing between sample groups, and for mapping splicing quantitative trait loci (sQTLs). Compared to contemporary methods, we find 1.4–2.1 times more sQTLs, many of which help us ascribe molecular effects to disease-associated variants. Strikingly, transcriptome-wide associations between LeafCutter intron quantifications and 40 complex traits increased the number of associated disease genes at 5% FDR by an average of 2.1-fold as compared to using gene expression levels alone. LeafCutter is fast, scalable, easy to use, and available online.

Patterns of resource selection by animal populations emerge as a result of the behavior of many individuals. Statistical models that describe these population-level patterns of habitat use can miss important interactions between individual animals and characteristics of their local environment; however, identifying these interactions is difficult. One approach to this problem is to incorporate models of individual movement into resource selection models. To do this, we propose a model for step selection functions (SSF) that is composed of a resource-independent movement kernel and a resource selection function (RSF). We show that standard case-control logistic regression may be used to fit the SSF; however, the sampling scheme used to generate control points (i.e., the definition of availability) must be accommodated. We used three sampling schemes to analyze simulated movement data and found that ignoring sampling and the resource-independent movement kernel yielded biased estimates of selection. The level of bias depended on the method used to generate control locations, the strength of selection, and the spatial scale of the resource map. Using empirical or parametric methods to sample control locations produced biased estimates under stronger selection; however, we show that the addition of a distance function to the analysis substantially reduced that bias. Assuming a uniform availability within a fixed buffer yielded strongly biased selection estimates that could be corrected by including the distance function but remained inefficient relative to the empirical and parametric sampling methods. As a case study, we used location data collected from elk in Yellowstone National Park, USA, to show that selection and bias may be temporally variable. Because under constant selection the amount of bias depends on the scale at which a resource is distributed in the landscape, we suggest that distance always be included as a covariate in SSF analyses. This approach to modeling resource selection is easily implemented using common statistical tools and promises to provide deeper insight into the movement ecology of animals.

Integration of genome-wide association studies (GWAS) and expression quantitative trait loci (eQTL) studies is needed to improve our understanding of the biological mechanisms underlying GWAS hits, and our ability to identify therapeutic targets. Gene-level association methods such as PrediXcan can prioritize candidate targets. However, limited eQTL sample sizes and absence of relevant developmental and disease context restrict our ability to detect associations. Here we propose an efficient statistical method (MultiXcan) that leverages the substantial sharing of eQTLs across tissues and contexts to improve our ability to identify potential target genes. MultiXcan integrates evidence across multiple panels using multivariate regression, which naturally takes into account the correlation structure. We apply our method to simulated and real traits from the UK Biobank and show that, in realistic settings, we can detect a larger set of significantly associated genes than using each panel separately. To improve applicability, we developed a summary result-based extension called S-MultiXcan, which we show yields highly concordant results with the individual level version when LD is well matched. Our multivariate model-based approach allowed us to use the individual level results as a gold standard to calibrate and develop a robust implementation of the summary-based extension. Results from our analysis as well as software and necessary resources to apply our method are publicly available.

Large-scale genomic and transcriptomic initiatives offer unprecedented insight into complex traits, but clinical translation remains limited by variant-level associations without biological context and lack of analytic resources. Our resource, PhenomeXcan, synthesizes 8.87 million variants from genome-wide association study summary statistics on 4091 traits with transcriptomic data from 49 tissues in Genotype-Tissue Expression v8 into a gene-based, queryable platform including 22,515 genes. We developed a novel Bayesian colocalization method, fast enrichment estimation aided colocalization analysis (fastENLOC), to prioritize likely causal gene-trait associations. We successfully replicate associations from the phenome-wide association studies (PheWAS) catalog Online Mendelian Inheritance in Man, and an evidence-based curated gene list. Using PhenomeXcan results, we provide examples of novel and underreported genome-to-phenome associations, complex gene-trait clusters, shared causal genes between common and rare diseases via further integration of PhenomeXcan with ClinVar, and potential therapeutic targets. PhenomeXcan (phenomexcan.org) provides broad, user-friendly access to complex data for translational researchers.

Understanding the genetic architecture of gene expression traits is key to elucidating the underlying mechanisms of complex traits. Here, for the first time, we perform a systematic survey of the heritability and the distribution of effect sizes across all representative tissues in the human body. We find that local h2 can be relatively well characterized with 59% of expressed genes showing significant h2 (FDR < 0.1) in the DGN whole blood cohort. However, current sample sizes (n ≤ 922) do not allow us to compute distal h2. Bayesian Sparse Linear Mixed Model (BSLMM) analysis provides strong evidence that the genetic contribution to local expression traits is dominated by a handful of genetic variants rather than by the collective contribution of a large number of variants each of modest size. In other words, the local architecture of gene expression traits is sparse rather than polygenic across all 40 tissues (from DGN and GTEx) examined. This result is confirmed by the sparsity of optimal performing gene expression predictors via elastic net modeling. To further explore the tissue context specificity, we decompose the expression traits into cross-tissue and tissue-specific components using a novel Orthogonal Tissue Decomposition (OTD) approach. Through a series of simulations we show that the cross-tissue and tissue-specific components are identifiable via OTD. Heritability and sparsity estimates of these derived expression phenotypes show similar characteristics to the original traits. Consistent properties relative to prior GTEx multi-tissue analysis results suggest that these traits reflect the expected biology. Finally, we apply this knowledge to develop prediction models of gene expression traits for all tissues. The prediction models, heritability, and prediction performance R2 for original and decomposed expression phenotypes are made publicly available (https://github.com/hakyimlab/PrediXcan).

We sought to comprehensively and systematically characterize the relationship between genetic variation, miRNA expression, and mRNA expression. Genome-wide expression profiling of samples of European and African ancestry identified in each population hundreds of miRNAs whose increased expression is correlated with correspondingly reduced expression of target mRNAs. We scanned 3' UTR SNPs with a potential functional effect on miRNA binding for cis-acting expression quantitative trait loci (eQTLs) for the corresponding proximal target genes. To extend sequence-based, localized analyses of SNP effect on miRNA binding, we proceeded to dissect the genetic basis of miRNA expression variation; we mapped miRNA expression levels-as quantitative traits-to loci in the genome as miRNA eQTLs, demonstrating that miRNA expression is under significant genetic control. We found that SNPs associated with miRNA expression are significantly enriched with those SNPs already shown to be associated with mRNA. Moreover, we discovered that many of the miRNA-associated genetic variations identified in our study are associated with a broad spectrum of human complex traits from the National Human Genome Research Institute catalog of published genome-wide association studies. Experimentally, we replicated miRNA-induced mRNA expression inhibition and the cis-eQTL relationship to the target gene for several identified relationships among SNPs, miRNAs, and mRNAs in an independent set of samples; furthermore, we conducted miRNA overexpression and inhibition experiments to functionally validate the miRNA-mRNA relationships. This study extends our understanding of the genetic regulation of the transcriptome and suggests that genetic variation might underlie observed relationships between miRNAs and mRNAs more commonly than has previously been appreciated.