Boyang Fu

Generative modeling of spatially resolved perturbation responses with CONCERT

CONCERT, a niche-aware generative model, predicts spatial perturbation responses more accurately than existing methods across multiple tissues and experimental setups.

Comprehensive gene heritability estimation reveals the genetic architecture of rare coding variants underlying complex traits

This paper introduces FLEX, a method for estimating gene-level heritability from whole-exome sequencing data, showing rare coding variants contribute significantly to complex trait heritability.

Investigating the sources of variable impact of pathogenic variants in monogenic metabolic conditions

The paper investigates why carriers of pathogenic variants in monogenic metabolic conditions show variable disease severity, finding support for variant effect sizes, polygenic backgrounds, and epistasis.

A Scalable Adaptive Quadratic Kernel Method for Interpretable Epistasis Analysis in Complex Traits

The paper introduces QuadKAST, a scalable algorithm for detecting epistasis in complex traits, and identifies 32 trait-gene pairs with significant quadratic effects in UK Biobank data.

A biobank-scale test of marginal epistasis reveals genome-wide signals of polygenic epistasis

The paper tests for polygenic epistasis in biobank-scale data and finds 16 trait-loci pairs with strong evidence of interaction effects.

Fast Kernel-based Association Testing of non-linear genetic effects for Biobank-scale data

FastKAST detects non-linear genetic effects on traits in large datasets by providing efficient kernel-based tests with calibrated hypothesis testing.

Leveraging family data to design Mendelian Randomization that is provably robust to population stratification

This paper introduces MR-Twin, a Mendelian Randomization method robust to population stratification, and shows it reduces false positives compared to standard methods in UK Biobank data.

GPC: An expressive and tractable deep generative model for genetic variation data

This paper introduces GPC, a deep generative model for genetic variation data that improves imputation accuracy and privacy preservation compared to existing methods.