Transancestral fine-mapping of four type 2 diabetes susceptibility loci highlights potential causal regulatory mechanisms

Momoko Horikoshi, Lorenzo Pasquali, Steven Wiltshire, Jeroen R. Huyghe, Anubha Mahajan, Jennifer L. Asimit, Teresa Ferreira, Adam E. Locke, Neil R. Robertson, Xu Wang, Xueling Sim, Hayato Fujita, Kazuo Hara, Robin Young, Weihua Zhang, Sungkyoung Choi, Han Chen, Ismeet Kaur, Fumihiko Takeuchi, Pierre Fontanillas, Dorothée Thuillier, Loic Yengo, Jennifer E. Below, Claudia H.T. Tam, Ying Wu, Gonçalo Abecasis, David Altshuler, Graeme I. Bell, John Blangero, Noél P. Burtt, Ravindranath Duggirala, Jose C. Florez, Craig L. Hanis, Mark Seielstad, Gil Atzmon, Juliana C.N. Chan, Ronald C.W., Philippe Froguel, James G. Wilson, Dwaipayan Bharadwaj, Josee Dupuis, James B. Meigs, Yoon Shin Cho, Taesung Park, Jaspal S. Kooner, John C. Chambers, Danish Saleheen, Takashi Kadowaki, E. Shyong Tai, Karen L. Mohlke, Nancy J. Cox, Jorge Ferrer, Eleftheria Zeggini, Norihiro Kato, Yik Ying Teo, Michael Boehnke, Mark I. McCarthy, Andrew P. Morris

To gain insight into potential regulatory mechanisms through which the effects of variants at four established type 2 diabetes (T2D) susceptibility loci (CDKAL1, CDKN2A-B, IGF2BP2 and KCNQ1) are mediated, we undertook transancestral fine-mapping in 22,086 cases and 42,539 controls of East Asian, European, South Asian, African American, and Mexican American descent. Through high-density imputation and conditional analyses, we identified seven distinct association signals at these four loci, each with allelic effects on T2D susceptibility that were homogenous across ancestry groups. By leveraging differences in the structure of linkage disequilibrium between diverse populations, and increased sample size, we localised the variants most likely to drive each distinct association signal. We demonstrated that integration of these genetic fine-mapping data with genomic annotation can highlight potential causal regulatory elements in T2D-relevant tissues. These analyses provide insight into the mechanisms through which T2D association signals are mediated, and suggest future routes to understanding the biology of specific disease susceptibility loci.