Gad Abraham

This study evaluates deep learning methods for polygenic prediction, showing that MLP benefits most from larger sample sizes and one-hot encoding improves performance across methods.

This study evaluates a polygenic risk score for coronary heart disease in individuals aged 70 and older, finding it improves prediction when added to conventional risk factors.

The paper uses polygenic scores to identify plasma proteins causally linked to cardiometabolic diseases and finds some are targets for existing therapies.

PGS Catalog standardizes 192 curated polygenic scores and metadata (86 traits); reapplied in UK Biobank, eight of nine colorectal cancer scores predicted disease in European-ancestry participants.

The paper develops a genomic risk score (metaGRS) for ischaemic stroke that outperforms existing risk factors in predicting stroke risk.

This study identifies genetic variants with pleiotropic effects on cytokines and related traits using a multivariate genome-wide association analysis.

The paper identifies alpha-1 antitrypsin as a key contributor to GlycA's predictive power for disease risk and mortality across multiple conditions.

This study evaluates how adding polygenic risk scores and clinical biochemistry markers to conventional CVD risk models improves prediction accuracy and informs statin therapy decisions in a UK primary care setting.

This paper develops a genomic risk score for coronary artery disease using 1.7 million variants and tests it in UK Biobank, showing it outperforms conventional risk factors in predicting CAD risk.

The paper integrates transcriptomic, metabolomic, and genomic data to identify gene networks and their genetic regulation linked to immune function and circulating metabolites.

The paper identifies genetic loci associated with coronary artery disease that show evidence of selection and antagonistic pleiotropy with reproductive traits.

FlashPCA2 performs PCA on large genotype datasets faster and with less memory than existing tools.

This study develops a genomic risk score for coronary heart disease and shows it improves risk prediction beyond clinical scores, particularly in older adults.

FlashPCA provides a fast implementation of sparse canonical correlation analysis for genomic data, enabling rapid analysis of large SNP and phenotype datasets.

The paper introduces NetRep, a permutation-based method to assess gene coexpression module preservation across tissues, finding 52% of modules are preserved in at least one tissue and 25% in all four tissues.

This study identifies epistatic interactions within the MHC region that contribute to celiac disease risk, finding 20 significant signals independent of known HLA haplotypes.

This paper develops a genomic risk score for celiac disease targeting HLA-DQ2.5-positive individuals, showing improved predictive performance over existing methods and reducing unnecessary follow-up tests.

This study links elevated glycoprotein acetyls (GlycA) to chronic inflammation and microbial infection risk using integrative omics and health records.

This paper reviews integrative omics approaches to uncover the genetic basis of coronary artery disease and improve disease prediction and treatment response.

The paper introduces flashpca, a fast PCA tool for large SNP datasets, achieving similar accuracy to existing methods but significantly reducing computation time.