Welcome to #StatFunGen Lab wiki

This wiki site is developed to share useful information with our lab members and collaborators.

Our Laboratory of Statistical Functional Genomics (#StatFunGen) in the Gertrude H. Sergievsky Center at Columbia University aims to understand molecular mechanisms of complex traits biology, using Alzheimer’s disease as a model to study how genetic variation shapes gene regulation across the central dogma and how dysregulation of these processes contributes to neurodegeneration. We approach this as a data science problem, “gather good information” (biotechnology and bioinformatics), “analyze information” (statistical modeling and inference), and “draw conclusions” (neurological sciences). Our work centers on molecular quantitative trait loci (QTL) from population-scale omics data as a path into studying genetic regulation, and we develop new computational methods as challenges arise from the data.

We work closely with collaborators at Columbia University, Rush University, and Washington University in St. Louis to generate and harmonize multi-omics data from brain tissues across multiple ancestries and from cerebrospinal fluid samples. More recently, through collaboration with the New York Genome Center and the PARDoS cohort, we are expanding these resources to study molecular changes in aging beyond Alzheimer’s disease. These efforts yield measurements spanning chromatin accessibility, histone modifications, DNA methylation, gene expression, alternative splicing, alternative polyadenylation, protein abundance, as well as emerging modalities such as m6A and pseudouridine modifications, glycoproteomics, lipidomics, and metabolomics.

We develop and apply statistical methods and computational pipelines to extract biological signal from these high-dimensional data, with particular emphasis on context-dependent causal effects across cell types, age groups, and populations. Our methodological contributions include genetic fine-mapping and colocalization to identify putative causal variants, and approaches to quantify new molecular phenotypes and define QTL particularly for histone acetylation, DNA methylation, chromatin accessibility, alternative splicing and gene modules. By integrating QTL data across molecular modalities with disease genetics, we address challenges arising from complex linkage disequilibrium, allelic heterogeneity, and the need to distinguish causal genes from correlated hitchhikers. We also leverage ancestral diversity to improve fine-mapping resolution and prediction of molecular phenotypes in underrepresented populations.

We use functional genomics to generate testable hypotheses about disease mechanisms, working closely with domain experts to carefully interpret what the data reveal. Specifically, we seek to integrate convergent evidence across molecular contexts, intermediate phenotypes such as neuroimaging endophenotypes, and disease endpoints to understand how genetic variation propagates through molecular layers to influence neurodegeneration. We integrate brain and CNS QTL data with epigenomic profiles derived from experimental assays and computational predictions to prioritize causal genes and characterize their cell-type-specific roles, paying particular attention to context-dependent effects that reveal compensatory mechanisms, stage-dependent vulnerabilities, and potential intervention points particularly informed by non-coding and/or rare variants.

Besides research, we are committed to developing resources that benefit the broader scientific community. We build and maintain software packages, bioinformatics pipelines, and curated data resources, both for methods we develop and for existing approaches we use extensively ourselves. We also invest in education and outreach, contributing to tutorials, lectures, and collaborative training efforts that strengthen the functional genomics research community.

We nurture a lab environment that values curiosity, creativity, and genuine enjoyment of the work. If you are interested in joining us to advance scientific research and your own growth, feel free to reach out to Gao Wang for information on possible openings.