Christoph Lippert

The paper introduces MIFM, a deep learning method to predict causal regulatory variants from GWAS data using weak labels and DNA sequences, improving cross-ancestry risk prediction and variant disentanglement.

TorchLIMIX accelerates multivariate GWAS using GPU computing, finding 37 additional SNPs in Arabidopsis metabolic traits compared to univariate GWAS.

This paper applies TransferGWAS to UK Biobank brain MRI data using deep features, identifying 289 loci and novel genetic correlations with traits like bone density and type 2 diabetes.

This study evaluates polygenic scoring methods across five biobanks and finds that biobank differences outweigh method differences, with ensemble learning improving performance.

HAPNEST generates and evaluates large-scale synthetic genotype and phenotype datasets for polygenic risk score analysis.

The paper performs a rare variant association study of 30 blood biomarkers in 191,971 UK Biobank individuals, identifying 193 gene-biomarker associations using functionally informed kernel-based tests.

transferGWAS applies deep transfer learning to medical images for GWAS, identifying 60 genomic regions associated with retinal fundus images, including 7 novel eye-related loci.

HistoFlow enables fast, label-efficient training of deep learning models for cell segmentation and analysis using interactive tools and artificial data.

The paper addresses criticisms of a previous study on identifying individuals from whole-genome sequencing data by demonstrating the limitations of Erlich's simulation and highlighting the algorithm's use of multiple trait-based predictive models.