Matthew R. Robinson

The paper introduces vampW, a Bayesian method for joint variable selection in omic biomarkers and time-to-event data, showing reduced false discoveries and improved prediction.

The paper introduces CI-GWAS, a method to improve causal inference in GWAS by addressing confounding and pleiotropy, finding few suitable IVs and replicating some but not all previously reported causal links.

The paper introduces JODIE, a model to estimate direct, indirect, and parent-of-origin genetic effects on height, BMI, and educational attainment using family trios from two biobanks.

This study identifies 58 genomic regions associated with meiotic recombination, 24 of which are sex-specific, and finds a moderate genetic correlation between male and female recombination rates.

This paper develops a quantitative genetic model to analyze how indirect genetic effects and genomic imprinting influence genetic variance under random and assortative mating.

The paper reassesses selection on height using GWAS and allele frequency data, finding positive selection at height-associated loci in Europe.

The paper uses graphical modeling on electronic health records to separate genetic effects on disease, treatment, and treatment response in large-scale biobank data.

The paper introduces MAJA, a method for joint association testing in multi-omics data, finding shared epigenetic pathways across multiple traits in methylation data.

This paper develops an epigenetic score to objectively measure alcohol consumption using blood-based DNA methylation and validates it across multiple cohorts, finding associations with brain health metrics.

This paper introduces gVAMP, an algorithm for analyzing whole genome sequence data to identify genetic variants associated with complex traits like height, showing improved accuracy over existing methods.

The paper identifies age-specific genetic effects on age at natural menopause using UK Biobank data, revealing varying biological pathways and health outcomes across menopause timing.

The paper refines epigenetic predictors of chronological and biological age using large-scale EWAS and external validation, improving cAge prediction and outperforming existing bAge predictors.

The paper estimates liability-scale heritability for low prevalence diseases using biobank data and finds differences in heritability estimates across theoretical methods.

This paper improves genomic prediction and GWAS discovery in biobank data using a Bayesian grouped mixture of regressions model, achieving higher accuracy and detecting more loci than existing methods.

This paper improves cancer genomic prediction by analyzing age-at-diagnosis instead of case-control data, finding new genomic regions and enhancing prediction accuracy using Bayesian time-to-event modeling.

This study identifies DNA methylation predictors of cognitive function using blood samples, showing they explain variance in cognitive ability and associate with biological markers.

The paper identifies genetic and epigenetic factors influencing plasma protein levels linked to Alzheimer's disease, finding novel associations with TREM2 and TBCA.

The paper introduces a Bayesian model, BayesRR-RC, to infer the genetic architecture of complex traits and finds that most genetic variance is attributed to non-proximal regulatory and coding regions.

This paper introduces BayesW, a Bayesian method for analyzing time-to-event phenotypes, showing improved statistical power and prediction accuracy for age-at-onset traits in large-scale datasets.

This study identifies genetic and epigenetic factors associated with inflammatory protein levels in older adults using GWAS and EWAS with a Bayesian framework.

SHAPEIT4 improves haplotype estimation with sub-linear scaling, high accuracy, and integration of external phasing information using UK Biobank and Genome In A Bottle datasets.

The paper quantifies assortative mating on complex traits by analyzing SNP data from 400,000 individuals, finding significant evidence for height and educational attainment.

This paper introduces a Bayesian method to estimate epigenetic associations with complex traits, finding that methylation data captures significant variance in BMI and smoking behavior.

The paper analyzes mitochondrial variation in the DGRP, identifies haplotype-specific phenotypic associations, and finds no major fitness effects from mito-nuclear incompatibilities.

The paper estimates the joint distribution of effect size and MAF for genetic variants in 28 complex traits and detects widespread negative selection signatures.

This study uses genetic data to identify causal relationships between risk factors and common diseases, finding protective and bidirectional effects.