Chris Wallace

Single-cell eQTL mapping in 5.4 million PBMCs from 1,925 individuals with matched WGS finds common-variant eQTLs for 83% and rare-variant signals for 47% of 21,404 genes.

The paper identifies immune cell type-specific gene expression regulated by tandem repeats using multi-omics data from large cohorts.

This study evaluates genomic foundation models' ability to improve common variant discovery in GWAS and finds they do not outperform standard functional annotations.

The paper evaluates eQTL-GWAS colocalisation studies, finding that 34-50% of GWAS hits colocalise, with trade-offs between granularity, sample size, and colocalisation detection.

This paper introduces a semi-supervised Bayesian method to integrate spatial proteomics data with other omic datasets to improve protein subcellular localization inference.

The paper benchmarks four colocalization methods using protein QTL data to assess their reliability and consistency in detecting shared genetic causes.

The paper shows that variable information across SNPs in GWAS data can lead to false colocalisation inferences and proposes a proportional colocalisation test to address this issue.

This paper introduces sfFDR to enhance variant discovery by integrating related trait summary statistics, increasing power and discovering additional SNPs in GWAS.

The paper introduces quasar, a software for eQTL mapping that offers faster performance and better statistical models compared to existing methods.

The paper maps cell-type-specific eQTLs in IBD patients to identify effector genes and mechanisms linking genetic risk to disease.

This study evaluates deep-learning methods for improving polygenic scores, finding limited benefits due to confounding effects from linkage disequilibrium.

The paper identifies MDA5 variants that protect against autoimmune diseases but increase IBD risk, showing a trade-off between antiviral activity and autoimmunity.

This paper compares methods for imputing cell-type specific expression from mixed RNA-seq data, finding that machine learning methods like LASSO and ridge outperform domain-specific methods in recovering differential gene expression.

The paper identifies trimodal QTLs that jointly affect enhancer accessibility, promoter contacts, and gene expression in monocytes, revealing genetic links between enhancer activity and chromatin boundaries.

This paper identifies common-variant associations with selective IgA deficiency using GWAS meta-analysis and pleiotropy-based methods, finding 26 variants outside the MHC and highlighting polygenic and immune-related biology.

This paper introduces DPMUnc, a Bayesian clustering method that incorporates data uncertainty, and applies it to GWAS and gene expression data to identify disease subgroups.

This paper introduces variant-specific priors in the coloc method to improve colocalisation analysis between GWAS and eQTL associations, showing improved performance and altered significance in some loci.

ShaPRS improves polygenic score accuracy by leveraging shared genetic effects across traits or ancestries, enhancing performance for complex diseases and across populations.

The paper introduces a nonparametric test for detecting genetic similarity between traits using GWAS summary statistics, showing improved performance in small-sample contexts compared to existing methods.

CoPheScan improves PheWAS by accounting for linkage disequilibrium, identifying causal associations and reducing false positives in genetic variant analyses.

BBmix improves genotype calling from RNA-sequencing by modeling RNA-specific biases, achieving higher accuracy in heterozygous calls compared to existing methods.

The paper maps promoter interactions in human ILC3s using PCHi-C and links GWAS risk variants to target genes, identifying new autoimmune disease risk genes.

This study finds that sharing GWAS summary statistics leads to about 75% more citations, regardless of journal impact factor, encouraging authors to share data in repositories like the GWAS catalog.

The paper introduces an R package called fcfdr that extends the cFDR framework to use binary functional genomic data in GWAS, improving power while controlling FDR, and applies it to type 1 diabetes.

The paper introduces a consensus clustering method for Bayesian mixture models, showing it can uncover target clustering structures and handle large datasets efficiently.

This paper introduces a semi-supervised Bayesian mixture model for joint batch correction and classification, showing improved performance on low-dimensional data compared to existing methods.

Flashfm improves fine-mapping of multiple quantitative traits by sharing information between traits, reducing potential causal variants by 20% in a Ugandan cohort.

This paper improves colocalisation analysis by integrating the SuSiE framework, enhancing accuracy when multiple causal variants are present.

RapidoPGS computes polygenic scores using only summary GWAS data without needing test datasets, achieving faster performance than some existing methods.

EPISPOT improves molecular QTL mapping by integrating epigenetic annotations to detect and interpret genetic hotspots with greater statistical power and biological insight.