Summary
(How) Can we draw causal conclusions from observational data?
My research focus is the development of novel methodology for causal inference and its application in life science and health data research. A particular focus of my research is the use of genetic variation as instrumental variables, a study design also known as Mendelian randomization. More broadly, my research interest is at the intersection of causal inference, high-dimensional statistics, machine-learning and statistical genetics. Relevant statistical methodology for my research includes instrumental variables, Bayesian causal structure learning, causal networks including machine learning approaches for DAG learning, doubly robust estimation, and counterfactuals.
Collaborations are an important aspect of my work, I am fortunate to work with experts in cardiovascular health, cancer, brain diseases in particular neurodegeneration and mental health.
For an informal discussion regarding Summer projects please get in touch!
Github: https://github.com/verena-zuber
Publications
Journals
Huang L, Tang S, Rietkerk J, et al., 2023, Polygenic analyses show important differences between MDD symptoms collected using PHQ9 and CIDI-SF., Medrxiv
Huang J, Su B, Karhunen V, et al., 2023, Inflammatory diseases, inflammatory biomarkers, and Alzheimer disease: an observational analysis and mendelian randomization, Neurology, Vol:100, ISSN:0028-3878, Pages:e568-e581
Desai R, John A, Saunders R, et al., 2023, Examining the Lancet Commission risk factors for dementia using Mendelian randomisation., Bmj Mental Health, Vol:26, ISSN:2755-9734, Pages:1-8
Karageorgiou V, Gill D, Bowden J, et al., 2022, Sparse Dimensionality Reduction Approaches in Mendelian Randomization with highly correlated exposures
Roychowdhury T, Klarin D, Levin MG, et al., 2022, Multi-ancestry GWAS deciphers genetic architecture of abdominal aortic aneurysm and highlights<i>PCSK9</i>as a therapeutic target