Instrumental variables
Two-stage least squares and related designs for estimating causal effects from observational evidence.
CAUSAL INFERENCE · STATISTICAL ROBUSTNESS · FIRST AUTHOR
Causal inference, measurement, and research leadership.
“GWAS advancements to investigate disease associations and biological mechanisms”
Clinical and Translational Discovery · Volume 4, Issue 3 · e296 · First published 1 May 2024
01 / QUANTITATIVE METHODS
Methods for separating association from evidence of causation.
The article reviews instrumental-variable and Mendelian-randomization approaches, robustness estimators, bias mechanisms, and measurement technologies used across disease-association research.
Robustness estimators
Inverse-variance weighting, Egger regression, Bayesian methods, mixture models, and block jackknife resampling.
Bias mechanisms
Confounding, reverse causation, multiple testing, and overlapping-sample contamination.
Measurement trade-offs
Coverage, sensitivity, and cost comparisons across sequencing, array-based, and lower-coverage technologies.
02 / LEADERSHIP & AUTHORSHIP
Research leadership across disciplines.
At Rutgers Institute for Health, I directed a seven-member team spanning bioscience, software, project management, and pre-med research contributors.
My role included setting the research scope, dividing technical work across disciplines, reviewing contributions, controlling the final analytical standard, and leading manuscript revision through acceptance.
The work required comparing estimators, identifying failure modes in observational inference, reconciling evidence across more than 150 studies, and communicating technical conclusions for expert review.
03 / CITATION
Omidiran, O., Patel, A., Usman, S., Mhatre, I., Abdelhalim, H., DeGroat, W., Narayanan, R., Singh, K., Mendhe, D., & Ahmed, Z. (2024). “GWAS advancements to investigate disease associations and biological mechanisms.” Clinical and Translational Discovery, 4(3), e296. https://doi.org/10.1002/ctd2.296