Biology subjects

Liley, J.

Publications and source records attributed to Liley, J..

5 recordsLinked to original sources

bioRxiv2021

fcfdr: an R package to leverage continuous and binary functional genomic data in GWAS

SummaryGWAS discovery is limited in power to detect associations that exceed the stringent genome-wide significance threshold, but this limitation can be alleviated by leveraging relevant auxiliary data. Frameworks utilising the conditional false discovery rate (cFDR) can be used to leverage continuous auxiliary data (including GWAS and functional genomic data) with GWAS test statistics and have been shown to increase power for GWAS discovery whilst controlling the FDR. Here, we describe an extension to the cFDR framework for binary auxiliary data (such as whether SNPs reside in regions of the genome with specific activity states) and introduce an all-encompassing R package to implement the cFDR approach, fcfdr, demonstrating its utility in an application to type 1 diabetes. Availability and implementationThe fcfdr R package is freely available at: https://github.com/annahutch/fcfdr. Scripts and data to reproduce the analysis in this paper are freely available at: https://annahutch.github.io/fcfdr/articles/t1d_app.html

genetics↗

bioRxiv2020

Genome-wide scale analyses identify novel BMI genotype-environment interactions using a conditional false discovery rate

Genotype-environment interaction (GxE) studies typically focus on variants with previously known marginal associations. While such two-step filtering greatly reduces the multiple testing burden, it can miss loci with pronounced GxE effects, which tend to have weaker marginal associations. To test for GxE effects on a genome-wide scale whilst leveraging information from marginal associations in a flexible manner, we combine the conditional false discovery rate with interaction test results obtained from StructLMM. After validating our approach, we applied this strategy to UK Biobank (UKBB) data to probe for GxE effects on BMI. Using 126,077 UKBB individuals for discovery, we identified known (FTO, MC4R, SEC16B) and novel GxE signals, many of which replicated (FAM150B/ALKAL2,TMEM18, EFR3B, ZNF596-FAM87A, LIN7C-BDNF, FAIM2, UNC79, LAT) in an independent subset of UKBB (n=126,076). Finally, when analysing the full UKBB cohort, we identified 140 candidate loci with GxE effects, highlighting the advantages of our approach.

genetics↗

bioRxiv2018

Genetically distinct clinical subsets, and associations with asthma and eosinophil abundance, within Eosinophilic Granulomatosis with Polyangiitis

Eosinophilic granulomatosis with polyangiitis (EGPA: formerly Churg-Strauss syndrome) is a rare inflammatory disease of unknown cause. 30% of patients have anti-neutrophil cytoplasm antibodies (ANCA) specific for myeloperoxidase (MPO). We performed a genome-wide association study (GWAS) of EGPA, testing 7.5 million genetic variants in 684 cases and 6,838 controls. Case-control analyses were performed for EGPA as a whole, and stratified by ANCA. To increase power, we used a conditional false discovery rate method to leverage findings from GWASs of related phenotypes. In total, 11 variants were associated with EGPA, two specifically with ANCA-negative EGPA, and one (HLA-DQ) with MPO+ANCA EGPA. Many variants were associated with asthma, eosinophilic and immune-mediated diseases and, strikingly, nine were associated with eosinophil count in the general population. Through Mendelian randomisation, we show that a primary tendency to eosinophilia underlies EGPA susceptibility. We demonstrate that EGPA comprises two genetically and clinically distinct syndromes, with ANCA-negative EGPA genetically more similar to asthma. MPO+ ANCA EGPA is an eosinophilic autoimmune disease sharing certain clinical features and an MHC association with MPO+ ANCA-associated vasculitis, while ANCA-negative EGPA may instead have a mucosal/barrier dysfunction origin. Five identified candidate genes are targets of therapies in development, supporting their exploration in EGPA.

genetics↗

bioRxiv2018

Accurate error control in high dimensional association testing using conditional false discovery rates

High-dimensional hypothesis testing is ubiquitous in the biomedical sciences, and informative covariates may be employed to improve power. The conditional false discovery rate (cFDR) is widely-used approach suited to the setting where the covariate is a set of p-values for the equivalent hypotheses for a second trait. Although related to the Benjamini-Hochberg procedure, it does not permit any easy control of type-1 error rate, and existing methods are over-conservative. We propose a new method for type-1 error rate control based on identifying mappings from the unit square to the unit interval defined by the estimated cFDR, and splitting observations so that each map is independent of the observations it is used to test. We also propose an adjustment to the existing cFDR estimator which further improves power. We show by simulation that the new method more than doubles potential improvement in power over unconditional analyses compared to existing methods. We demonstrate our method on transcriptome-wide association studies, and show that the method can be used in an iterative way, enabling the use of multiple covariates successively. Our methods substantially improve the power and applicability of cFDR analysis.

genomics↗

bioRxiv2018

Improved consistency in estimates of conditional false discovery rates increases power relative to both existing methods and parametric estimators

A common aim in high-dimensional association studies is the identification of the subset of investigated variables associated with a trait of interest. Using association statistics on the same variables for a second related trait can improve power. An important quantity in such analyses is the conditional false-discovery rate (cFDR), the probability of non-association with the trait of interest given p-value thresholds for both traits. The cFDR can be used for hypothesis testing and as a posterior probability in its own right. In this paper, we propose new estimators for the cFDR based on kernel density estimates and mixture-Gaussian models of effect sizes, the latter also allowing estimation of a local form of cFDR (cfdr). We also propose a general non-parametric improvement to existing estimators based on estimating a posterior probability previously estimated at 1. We find that new estimators have the desirable property of smooth rejection regions, but, unexpectedly, do not improve the power of the method, even when distributional assumptions are true. Furthermore, we find that although the local cfdr represents a theoretically optimal decision boundary, noisiness in its estimation means it is less powerful than corresponding cFDR estimates. We find, however, that the non-parametric adjustment increases power for every estimator. We demonstrate the best method on transcriptome-wide association study datasets for breast and ovarian cancers. The findings from this analysis are of both theoretical and pragmatic interest, giving insight into the nature of cFDR and the behaviour of false-discovery rates in a two-dimensional setting. Our methods allow improved control over the behaviour of the cFDR estimator and improved power in high-dimensional hypothesis testing.

genomics↗