Novel Variance-Component TWAS method for studying complex human diseases with applications to Alzheimer’s dementia

Abstract

Transcriptome-wide association studies (TWAS) have been widely used to integrate transcriptomic and genetic data to study complex human diseases. Within a test dataset lacking transcriptomic data, traditional two-stage TWAS methods first impute gene expression by creating a weighted sum that aggregates SNPs with their corresponding cis-eQTL effects on reference transcriptome. Traditional TWAS methods then employ a linear regression model to assess the association between imputed gene expression and test phenotype, thereby assuming the effect of a cis-eQTL SNP on test phenotype is a linear function of the eQTL’s estimated effect on reference transcriptome. To increase TWAS robustness to this assumption, we propose a novel Variance-Component TWAS procedure (VC-TWAS) that assumes the effects of cis-eQTL SNPs on phenotype are random (with variance proportional to corresponding reference cis-eQTL effects) rather than fixed. VC-TWAS is applicable to both continuous and dichotomous phenotypes, as well as individual-level and summary-level GWAS data. Using simulated data, we show VC-TWAS is more powerful than traditional TWAS methods based on a two-stage Burden test, especially when eQTL genetic effects on test phenotype are no longer a linear function of their eQTL genetic effects on reference transcriptome. We further applied VC-TWAS to both individual-level (N = ~3.4K) and summary-level (N = ~54K) GWAS data to study Alzheimer’s dementia (AD). With the individual-level data, we detected 13 significant risk genes including 6 known GWAS risk genes such as TOMM40 that were missed by traditional TWAS methods. With the summary-level data, we detected 57 significant risk genes considering only cis-SNPs and 71 significant genes considering both cis- and trans- SNPs; these findings also validated our findings with the individual-level GWAS data. Our VC-TWAS method is implemented in the TIGAR tool for public use.

On the interpretation of transcriptome-wide association studies

“…Note that some methods [20][21][22] diverge from this linear model structure (Table 1, non-linear models). Statistically these can be seen as generalizations of the TWAS framework, though conceptually they can no longer be interpreted as imputing the genetic component of gene expression.…”

TIGAR-V2: Efficient TWAS Tool with Nonparametric Bayesian eQTL Weights of 49 Tissue Types from GTEx V8

“…TIGAR-V2 implements the variance-component TWAS test by [15] using the sequence kernel association test (SKAT) framework [18] with variant weights provided by eQTL effect size estimates w. Variance-component TWAS test is recommended if the assumption of the linear relationship between the SNP effect sizes on phenotype and eQTL weights is violated (see Text S2.2). Note that here the eQTL weights w are specific to the test gene and specific to the tissue type of the reference transcriptomic data.…”

“…With summary-level GWAS data (i.e., Z-score statistic values from single variant GWAS tests) of test samples, TIGAR-V2 tests the gene-based association by using both burden [16,17] and variance-component [15] test statistics, where cis-eQTL effect size estimates w are taken as variant weights.…”

“…TIGAR-V2 can perform TWAS using either individual-level or summary-level GWAS data. Besides the burden type TWAS test [1], the software further implements an additional variance-component test [15] for TWAS that retains power under model misspecification.…”

“…recently published variance-component gene-based association test[15]. For example, for training gene expression prediction models by Bayesian DPR method with 129 samples, ∼ 1800 SNPs per gene, four genes, and a single core, the computation time is reduced by up to 90% and memory usage by up to 50%, compared to the initial TIGAR tool.…”