Variant-specific inflation factors for assessing population stratification at the phenotypic variance level
Tamar Sofer, Xiuwen Zheng, Cecelia Laurie, Stephanie M. Gogarten, Jennifer A. Brody, Matthew P. Conomos, Joshua C. Bis, Timothy A. Thornton, Adam A. Szpiro, Jeffrey R. O’Connell, Ethan M. Lange, Yan Gao, L. Adrienne Cupples, Bruce M. Psaty, Namiko Abe, Gonçalo R. Abecasis, François Aguet, Christine M. Albert, Laura Almasy, Álvaro Alonso, Seth A. Ament, Peter Anderson, Pramod Anugu, Deborah Applebaum‐Bowden, Kristin Ardlie, Dan Arking, Donna K. Arnett, Allison E. Ashley‐Koch, Stella Aslibekyan, Tim Assimes, Paul L. Auer, Dimitrios Avramopoulos, Najib Ayas, Adithya Balasubramanian, John Barnard, Kathleen C. Barnes, R. Graham Barr, Emily Barron‐Casella, Lucas Barwick, Terri Beaty, Gerald J. Beck, Diane M. Becker, Lewis C. Becker, Rebecca Beer, Amber L. Beitelshees, Emelia J. Benjamin, Takis Benos, Marcos Bezerra, Larry Bielak, Joshua C. Bis, Thomas W. Blackwell, John Blangero, Eric Boerwinkle, Donald W. Bowden, Russell P. Bowler, Jennifer A. Brody, Ulrich Broeckel, Jai Broome, Deborah Brown, Karen Bunting, Esteban G. Burchard, Carlos D. Bustamante, Erin Buth, Brian E. Cade, Jonathan Cardwell, Vincent J. Carey, Julie Carrier, Cara L. Carty, Richard Casaburi, Juan P. Romero, James F. Casella, Peter J. Castaldi, Mark Chaffin, Christy Chang, Yi‐Cheng Chang, Daniel I. Chasman, Sameer Chavan, Bo-Juen Chen, Wei‐Min Chen, Yii‐Der Ida Chen, Michael H. Cho, Seung Hoan Choi, Lee‐Ming Chuang, Mina K. Chung, Ren‐Hua Chung, Clary B. Clish, Suzy Comhair, Matthew P. Conomos, Elaine Cornell, Adolfo Correa, Carolyn Crandall, James D. Crapo, L. Adrienne Cupples, Joanne E. Curran, Jeffrey L. Curtis, Brian Custer, Coleen Damcott, Dawood Darbar, Sean P. David, Colleen Davis
Abstract
In modern Whole Genome Sequencing (WGS) epidemiological studies, participant-level data from multiple studies are often pooled and results are obtained from a single analysis. We consider the impact of differential phenotype variances by study, which we term 'variance stratification'. Unaccounted for, variance stratification can lead to both decreased statistical power, and increased false positives rates, depending on how allele frequencies, sample sizes, and phenotypic variances vary across the studies that are pooled. We develop a procedure to compute variant-specific inflation factors, and show how it can be used for diagnosis of genetic association analyses on pooled individual level data from multiple studies. We describe a WGS-appropriate analysis approach, implemented in freely-available software, which allows study-specific variances and thereby improves performance in practice. We illustrate the variance stratification problem, its solutions, and the proposed diagnostic procedure, in simulations and in data from the Trans-Omics for Precision Medicine Whole Genome Sequencing Program (TOPMed), used in association tests for hemoglobin concentrations and BMI.