| Title: | Leveraging eQTLs to Identify Individual-Level Tissue of Interest for a Complex Trait |
|---|---|
| Description: | Genetic predisposition for complex traits is often manifested through multiple tissues of interest at different time points in the development. As an example, the genetic predisposition for obesity could be manifested through inherited variants that control metabolism through regulation of genes expressed in the brain and/or through the control of fat storage in the adipose tissue by dysregulation of genes expressed in adipose tissue. We present a method eGST (eQTL-based genetic subtyper) that integrates tissue-specific eQTLs with GWAS data for a complex trait to probabilistically assign a tissue of interest to the phenotype of each individual in the study. eGST estimates the posterior probability that an individual's phenotype can be assigned to a tissue based on individual-level genotype data of tissue-specific eQTLs and marginal phenotype data in a genome-wide association study (GWAS) cohort. Under a Bayesian framework of mixture model, eGST employs a maximum a posteriori (MAP) expectation-maximization (EM) algorithm to estimate the tissue-specific posterior probability across individuals. Methodology is available from: A Majumdar, C Giambartolomei, N Cai, MK Freund, T Haldar, T Schwarz, J Flint, B Pasaniuc (2019) <doi:10.1101/674226>. |
| Authors: | Arunabha Majumdar [aut, cre], Tanushree Haldar [aut], Bogdan Pasaniuc [aut] |
| Maintainer: | Arunabha Majumdar <[email protected]> |
| License: | GPL-3 |
| Version: | 1.0.0 |
| Built: | 2024-11-08 04:12:37 UTC |
| Source: | https://github.com/arunabhacodes/egst |
Genetic predisposition for complex traits is often manifested through multiple tissues of interest at different time points in the development. As an example, the genetic predisposition for obesity could be manifested through inherited variants that control metabolism through regulation of genes expressed in the brain and/or through the control of fat storage in the adipose tissue by dysregulation of genes expressed in adipose tissue. We present a method eGST that integrates tissue-specific eQTLs with GWAS data for a complex trait to probabilistically assign a tissue of interest to the phenotype of each individual in the study. eGST estimates the posterior probability that an individual's phenotype can be assigned to a tissue based on individual-level genotype data of tissue-specific eQTLs and marginal phenotype data in a GWAS cohort. Under a Bayesian framework of mixture model, eGST employs a maximum a posteriori (MAP) expectation-maximization (EM) algorithm to estimate the tissue-specific posterior probability across individuals.
eGSTIt estimates the posterior probability that the genetic susceptibility of the phenotype of an individual in the study is mediated through eQTLs specific to a tissue of interest. The phenotype across individuals can be classified into tissues under consideration based on the estimated tissue-specific posterior probability across individuals.
Maintainer: Arunabha Majumdar [email protected]
Authors:
Tanushree Haldar [email protected]
Bogdan Pasaniuc [email protected]
Majumdar A, Giambartolomei C, Cai N, Freund MK, Haldar T, J Flint, Pasaniuc B (2019) Leveraging eQTLs to identify tissue-specific genetic subtype of complex trait. bioRxiv.
Useful links:
Run eGST to estimate the posterior probability that the genetic susceptibility of the phenotype of an individual in the study is mediated through eQTLs specific to a tissue of interest. To create sets of tissue-specific eQTLs in your context, please see our manuscript: Majumdar A, Giambartolomei C, Cai N, Freund MK, Haldar T, J Flint, Pasaniuc B (2019) Leveraging eQTLs to identify tissue-specific genetic subtype of complex trait, bioRxiv.
eGST(pheno, geno, tissues, logLimprovement = 5 * 10^(-8), seed_choice = sample(1:1000, size = 1), nIter = 100)eGST(pheno, geno, tissues, logLimprovement = 5 * 10^(-8), seed_choice = sample(1:1000, size = 1), nIter = 100)
pheno |
A numeric vector of length N where N is the number of individuals. It contains the GWAS phenotype values of individuals. No default. |
geno |
A list with K elements where K is the number of tissues. Each element of geno is the genotype matrix of the eQTLs specific to a tissue in the GWAS cohort. So j-th element of geno is N by Mj matrix containing the genotype data of N individuals (rows) at Mj eQTLs (columns) specific to j-th tissue. Each eQTL is a bi-allelic SNP with minor allele frequency > 0.01. Genotypes at each eQTL must be normalized across N individuals. If 0/1/2 valued genotype matrix is provided, it is internally normalized. No default. |
tissues |
A character vector of length K. It contains the names of tissues of interest in the analysis. The order of tissues in this vector must match the order of tissues in the previous argument 'geno'. No default. |
logLimprovement |
A positive real number specifying the minimum possible improvement
of data log-likelihood in MAP-EM stopping criterion. Default |
seed_choice |
An integer providing the choice of random seed for initialization in MAP-EM algorithm. Default is an integer randomly selected in (1,...,1000). |
nIter |
An integer providing the maximum number of iterations allowed in the MAP-EM algorithm. Default is 100. |
The output produced by eGST is a list which consists of various components.
gamma |
A N by K matrix providing the tissue-specific posterior probability of N individuals across K tissues. |
alfa |
Baseline tissue-specific intercepts/means of the trait. |
beta |
Tissue-specific eQTLs' genetic effect on the trait. |
sigma_g |
Square root of the variance of tissue-specific per-eQTL genetic effect on the trait. |
sigma_e |
Square root of the error variance of tissue-specific subtype of the trait which remains unexplained by the tissue-specific eQTLs. |
m |
Number of tissue-specific eQTLs. |
logL |
log-likelihood of the data. |
A Majumdar, C Giambartolomei, N Cai, MK Freund, T Haldar, T Schwarz, J Flint, B Pasaniuc (2019) Leveraging eQTLs to identify tissue-specific genetic subtype of complex trait. bioRxiv.
data(ExamplePhenoData) pheno <- ExamplePhenoData head(pheno) data(ExampleEQTLgenoData) geno <- ExampleEQTLgenoData geno[[1]][1:5,1:5] geno[[2]][1:5,1:5] tissues <- paste("tissue", 1:2, sep = "") result <- eGST(pheno, geno, tissues) str(result)data(ExamplePhenoData) pheno <- ExamplePhenoData head(pheno) data(ExampleEQTLgenoData) geno <- ExampleEQTLgenoData geno[[1]][1:5,1:5] geno[[2]][1:5,1:5] tissues <- paste("tissue", 1:2, sep = "") result <- eGST(pheno, geno, tissues) str(result)
ExampleEQTLgenoData is a list with two elements each containing a 1000 by 100 ordered genotype matrix. Each matrix provides the genotype data of the 1000 individuals at 100 tissue-specific eQTLs for each tissue.
data(ExampleEQTLgenoData)data(ExampleEQTLgenoData)
A list of two numeric matrix each having 1000 rows (individuals) and 100 columns (eQTLs):
data(ExampleEQTLgenoData) geno <- ExampleEQTLgenoDatadata(ExampleEQTLgenoData) geno <- ExampleEQTLgenoData
ExamplePhenoData is a simulated vector of phenotype for 1000 individuals. In this simulated example dataset, we have considered two tissues and corresponding sets of 100 tissue-specific eQTLs each. First half of 1000 individuals' phenotypes were simulated to have genetic effect from the first tissue specific eQTLs, but no effect from the second tissue-specific eQTLs. Hence first 500 individuals were assigned to the first tissue. Similarly, second half of the 1000 individuals were simulated to have genetic effect from the second-tissue specific eQTLs and hence they were assigned to the second tissue.
data(ExamplePhenoData)data(ExamplePhenoData)
A numeric vector of length 1000 containing the phenotype data of 1000 individuals.
data(ExamplePhenoData) pheno <- ExamplePhenoDatadata(ExamplePhenoData) pheno <- ExamplePhenoData