Farhad Hormozdiari

This paper introduces a deep learning framework, REGLE, that improves genomic discovery and risk prediction for respiratory and circulatory diseases by learning low-dimensional embeddings from high-dimensional clinical data.

This paper introduces M-REGLE, a multimodal deep learning method that improves genetic analysis of cardiovascular traits by integrating multiple clinical data modalities and outperforms unimodal approaches in locus discovery and risk prediction.

This study evaluates ML-based phenotyping's robustness to label noise using VCDR data, showing improved genetic discovery and PRS predictive power with denoising.

This paper trains a deep neural network on noisy COPD labels to predict COPD status from spirograms, improving genetic discovery by identifying new loci and replicating existing ones.

The paper develops a combined SNP-to-gene linking strategy (cS2G) to improve the identification of disease-causing genes and assess disease omnigenicity using GWAS data.

This paper introduces MsCAVIAR, a method for fine mapping causal variants across multiple GWAS studies, showing it reduces causal set sizes by over 20% in Type 2 Diabetes analysis.

The paper uses GTEx data to analyze GWAS loci, showing dose-dependent RNA effects and proposing target genes for 47% of loci using combined methods.

This paper introduces PolyFun, a method that improves fine-mapping accuracy by integrating genome-wide functional data, leading to more causal variant identifications and polygenic localization of trait heritability.

This study evaluates how well deep learning annotations predict disease heritability and finds they provide limited unique information beyond existing functional annotations.

The paper presents a large ASE resource from GTEx v8 with 431 million SNP-level and 153 million haplotype-level measurements, and extends phASER to estimate cis regulatory variant effect sizes.

The paper uses bulk RNA-seq data to estimate cell type abundances and identifies cell type interaction QTLs that provide finer resolution of tissue specificity than bulk cis-eQTLs.

The paper assesses the impact of admixture and ancestry on eQTL analysis and GWAS colocalization in GTEx, finding that local ancestry adjustment improves power but has limited effect on colocalization probabilities.

The paper examines telomere length variation across 6,391 tissue samples from 952 individuals, showing tissue-specific differences, positive correlations between tissues, and associations with age and ancestry.

Integrating expression, allelic expression and splicing across tissues identifies rare variants with relevance to traits.

The paper quantifies protein levels across 32 human tissues and compares them to transcriptome data, revealing insights into post-transcriptional regulation and disease mechanisms.

The paper constructs cell-type-specific TF binding annotations by combining sequence-based predictions with chromatin data, showing increased heritability enrichment for 49 diseases.

GTEx v8 maps expression and splicing QTLs in 49 tissues from 838 donors: cis-eQTLs cover 94.7% of protein-coding genes, and cross-tissue QTL sharing mirrors cell-type composition.

The paper analyzes 41 diseases and traits to show transposable elements contribute uniquely to disease heritability, with older elements and SINEs showing specific enrichment patterns.

The paper assesses gene network connectivity's role in disease heritability using LD score regression and finds that network annotations are enriched but largely explained by regulatory annotations.

The paper finds that disease heritability enrichment is higher in ancient, conserved regulatory elements compared to younger or less conserved ones.

This paper introduces MARS, a statistical method that improves GWAS power by considering multiple causal variants, and shows it identifies more eGenes and association loci than existing methods.

The paper shows that polygenicity of complex traits is explained by negative selection, which flattens genetic signal across the genome.

The paper introduces a method to detect directional effects of transcription factor binding on disease risk using GWAS data and applies it to identify new causal genes and mechanisms.

The paper uses molecular QTL annotations to identify genetic variants associated with disease heritability, finding stronger enrichments using causal posterior probabilities than standard annotations.

The paper develops S-PrediXcan to compute gene expression associations from GWAS summary data and applies it to 44 GTEx tissues and 100+ phenotypes, finding tissue-specific effects and colocalized associations.

This paper develops a statistical method that uses allele-specific expression and eQTL data to improve fine-mapping of causal variants for gene expression, achieving higher specificity than eQTL-only approaches.

This paper reviews how the multivariate normal distribution framework can be used to analyze GWAS by accounting for variant correlations, performing association testing, and enabling fine mapping and imputation.

The paper develops a method to infer allelic heterogeneity in GWAS and eQTL datasets, identifying 4152 loci with strong evidence of AH across different traits.

The paper identifies targets of selection in the FADS region in Europeans, showing allele frequency changes linked to lipid-related phenotypes and dietary adaptations.