GenoExp: a web tool for predicting gene expression levels from single nucleotide polymorphisms

A COVARIANCE-ENHANCED APPROACH TO MULTI-TISSUE JOINT EQTL MAPPING WITH APPLICATION TO TRANSCRIPTOME-WIDE ASSOCIATION STUDIES

Transcriptome-wide association studies based on genetically predicted gene expression have the potential to identify novel regions associated with various complex traits. It has been shown that incorporating expression quantitative trait loci (eQTLs) corresponding to multiple tissue types can improve power for association studies involving complex etiology. In this article, we propose a new multivariate response linear regression model and method for predicting gene expression in multiple tissues simultaneously. Unlike existing methods for multi-tissue joint eQTL mapping, our approach incorporates tissue-tissue expression correlation, which allows us to more efficiently handle missing expression measurements and more accurately predict gene expression using a weighted summation of eQTL genotypes. We show through simulation studies that our approach performs better than the existing methods in many scenarios. We use our method to estimate eQTL weights for 29 tissues collected by GTEx, and show that our approach significantly improves expression prediction accuracy compared to competitors. Using our eQTL weights, we perform a multi-tissue-based S-MultiXcan [2] transcriptome-wide association study and show that our method leads to more discoveries in novel regions and more discoveries overall than the existing methods. Estimated eQTL weights and code for implementing the method are available for download online at github.com/ajmolstad/MTeQTLResults.

Leveraging Methylation Alterations to Discover Potential Causal Genes Associated With the Survival Risk of Cervical Cancer in TCGA Through a Two-Stage Inference Approach

BACKGROUND

Multiple genes were previously identified to be associated with cervical cancer; however, the genetic architecture of cervical cancer remains unknown and many potential causal genes are yet to be discovered.

METHODS

To explore potential causal genes related to cervical cancer, a two-stage causal inference approach was proposed within the framework of Mendelian randomization, where the gene expression was treated as exposure, with methylations located within the promoter regions of genes serving as instrumental variables. Five prediction models were first utilized to characterize the relationship between the expression and methylations for each gene; then, the methylation-regulated gene expression (MReX) was obtained and the association was evaluated via Cox mixed-effect model based on MReX. We further implemented the aggregated Cauchy association test (ACAT) combination to take advantage of respective strengths of these prediction models while accounting for dependency among the p-values.

RESULTS

A total of 14 potential causal genes were discovered to be associated with the survival risk of cervical cancer in TCGA when the five prediction models were separately employed. The total number of potential causal genes was brought to 23 when conducting ACAT. Some of the newly discovered genes may be novel (e.g., YJEFN3, SPATA5L1, IMMP1L, C5orf55, PPIP5K2, ZNF330, CRYZL1, PPM1A, ESCO2, ZNF605, ZNF225, ZNF266, FICD, and OSTC). Functional analyses showed that these genes were enriched in tumor-associated pathways. Additionally, four genes (i.e., COL6A1, SYDE1, ESCO2, and GIPC1) were differentially expressed between tumor and normal tissues.

CONCLUSION

Our study discovered promising candidate genes that were causally associated with the survival risk of cervical cancer and thus provided new insights into the genetic etiology of cervical cancer.

Aggregating multiple expression prediction models improves the power of transcriptome-wide association studies

Transcriptome-wide association study (TWAS) is an important integrative method for identifying genes that are causally associated with phenotypes. A key step of TWAS involves the construction of expression prediction models for every gene in turn using its cis-SNPs as predictors. Different TWAS methods rely on different models for gene expression prediction, and each such model makes a distinct modeling assumption that is often suitable for a particular genetic architecture underlying expression. However, the genetic architectures underlying gene expression vary across genes throughout the transcriptome. Consequently, different TWAS methods may be beneficial in detecting genes with distinct genetic architectures. Here, we develop a new method, HMAT, which aggregates TWAS association evidence obtained across multiple gene expression prediction models by leveraging the harmonic mean P-value combination strategy. Because each expression prediction model is suited to capture a particular genetic architecture, aggregating TWAS associations across prediction models as in HMAT improves accurate expression prediction and enables subsequent powerful TWAS analysis across the transcriptome. A key feature of HMAT is its ability to accommodate the correlations among different TWAS test statistics and produce calibrated P-values after aggregation. Through numerical simulations, we illustrated the advantage of HMAT over commonly used TWAS methods as well as ad hoc P-value combination rules such as Fisher's method. We also applied HMAT to analyze summary statistics of nine common diseases. In the real data applications, HMAT was on average 30.6% more powerful compared to the next best method, detecting many new disease-associated genes that were otherwise not identified by existing TWAS approaches. In conclusion, HMAT represents a flexible and powerful TWAS method that enjoys robust performance across a range of genetic architectures underlying gene expression.

A deep auto-encoder model for gene expression prediction

BACKGROUND

Gene expression is a key intermediate level that genotypes lead to a particular trait. Gene expression is affected by various factors including genotypes of genetic variants. With an aim of delineating the genetic impact on gene expression, we build a deep auto-encoder model to assess how good genetic variants will contribute to gene expression changes. This new deep learning model is a regression-based predictive model based on the MultiLayer Perceptron and Stacked Denoising Auto-encoder (MLP-SAE). The model is trained using a stacked denoising auto-encoder for feature selection and a multilayer perceptron framework for backpropagation. We further improve the model by introducing dropout to prevent overfitting and improve performance.

RESULTS

To demonstrate the usage of this model, we apply MLP-SAE to a real genomic datasets with genotypes and gene expression profiles measured in yeast. Our results show that the MLP-SAE model with dropout outperforms other models including Lasso, Random Forests and the MLP-SAE model without dropout. Using the MLP-SAE model with dropout, we show that gene expression quantifications predicted by the model solely based on genotypes, align well with true gene expression patterns.

CONCLUSION

We provide a deep auto-encoder model for predicting gene expression from SNP genotypes. This study demonstrates that deep learning is appropriate for tackling another genomic problem, i.e., building predictive models to understand genotypes' contribution to gene expression. With the emerging availability of richer genomic data, we anticipate that deep learning models play a bigger role in modeling and interpreting genomics.

Cis-SNPs Set Testing and PrediXcan Analysis for Gene Expression Data using Linear Mixed Models

Understanding the functional mechanism of SNPs identified in GWAS on complex diseases is currently a challenging task. The studies of expression quantitative trait loci (eQTL) have shown that regulatory variants play a crucial role in the function of associated SNPs. Detecting significant genes (called eGenes) in eQTL studies and analyzing the effect sizes of cis-SNPs can offer important implications on the genetic architecture of associated SNPs and interpretations of the molecular basis of diseases. We applied linear mixed models (LMM) to the gene expression level and constructed likelihood ratio tests (LRT) to test for eGene in the Geuvadis data. We identified about 11% genes as eGenes in the Geuvadis data and found some eGenes were enriched in approximately independent linkage disequilibrium (LD) blocks (e.g. MHC). We further performed PrediXcan analysis for seven diseases in the WTCCC data with weights estimated using LMM and identified 64, 5, 21 and 1 significant genes (p < 0.05 after Bonferroni correction) associated with T1D, CD, RA and T2D. We found most of the significant genes of T1D and RA were also located within the MHC region. Our results provide strong evidence that gene expression plays an intermediate role for the associated variants in GWAS.

Prediction of gene expression with cis-SNPs using mixed models and regularization methods

Background

It has been shown that gene expression in human tissues is heritable, thus predicting gene expression using only SNPs becomes possible. The prediction of gene expression can offer important implications on the genetic architecture of individual functional associated SNPs and further interpretations of the molecular basis underlying human diseases.

Methods

We compared three types of methods for predicting gene expression using only cis-SNPs, including the polygenic model, i.e. linear mixed model (LMM), two sparse models, i.e. Lasso and elastic net (ENET), and the hybrid of LMM and sparse model, i.e. Bayesian sparse linear mixed model (BSLMM). The three kinds of prediction methods have very different assumptions of underlying genetic architectures. These methods were evaluated using simulations under various scenarios, and were applied to the Geuvadis gene expression data.

Results

The simulations showed that these four prediction methods (i.e. Lasso, ENET, LMM and BSLMM) behaved best when their respective modeling assumptions were satisfied, but BSLMM had a robust performance across a range of scenarios. According to R² of these models in the Geuvadis data, the four methods performed quite similarly. We did not observe any clustering or enrichment of predictive genes (defined as genes with R² ≥ 0.05) across the chromosomes, and also did not see there was any clear relationship between the proportion of the predictive genes and the proportion of genes in each chromosome. However, an interesting finding in the Geuvadis data was that highly predictive genes (e.g. R² ≥ 0.30) may have sparse genetic architectures since Lasso, ENET and BSLMM outperformed LMM for these genes; and this observation was validated in another gene expression data. We further showed that the predictive genes were enriched in approximately independent LD blocks.

Conclusions

Gene expression can be predicted with only cis-SNPs using well-developed prediction models and these predictive genes were enriched in some approximately independent LD blocks. The prediction of gene expression can shed some light on the functional interpretation for identified SNPs in GWASs.