TY - JOUR T1 - Evolutionary and functional data power search for obsessive-compulsive disorder genes JF - bioRxiv DO - 10.1101/107193 SP - 107193 AU - Hyun Ji Noh AU - Ruqi Tang AU - Jason Flannick AU - Colm O’Dushlaine AU - Ross Swofford AU - Daniel Howrigan AU - Diane P. Genereux AU - Jeremy Johnson AU - Gerard van Grootheest AU - Edna Grünblatt AU - Erik Andersson AU - Diana R. Djurfeldt AU - Paresh D. Patel AU - Michele Koltookian AU - Christina Hultman AU - Michele T. Pato AU - Carlos N. Pato AU - Steven A. Rasmussen AU - Michael A. Jenike AU - Gregory L. Hanna AU - S. Evelyn Stewart AU - James A. Knowles AU - Stephan Ruhrmann AU - Hans-Jörgen Grabe AU - Michael Wagner AU - Christian Rück AU - Carol A. Mathews AU - Susanne Walitza AU - Daniëlle C. Cath AU - Guoping Feng AU - Elinor K. Karlsson AU - Kerstin Lindblad-Toh Y1 - 2017/01/01 UR - http://biorxiv.org/content/early/2017/02/09/107193.abstract N2 - Obsessive-compulsive disorder (OCD) is a severe psychiatric disorder linked to abnormalities in the cortico-striatal circuit and in glutamate signaling. We sequenced coding and regulatory elements for 608 genes implicated in OCD from humans and two animal models (mouse and dog). Using a new method, PolyStrat, which prioritizes variants disrupting evolutionarily conserved, functional regions, we found four strongly associated genes when comparing 592 cases to 560 controls. These results were validated in a second, larger cohort. NRXN1 and HTR2A are enriched for coding variants altering postsynaptic protein-binding domains, while CTTNBP2 (synapse maintenance) and REEP3 (vesicle trafficking) are enriched for regulatory variants. The rare coding variant burden in NRXN1 achieves genomewide significance (p=6.37×10−11) when we include public data for 33,370 controls. Of 17 regulatory variants identified in CTTNBP2 and REEP3, we show that at least six alter transcription factor-DNA binding in human neuroblastoma cells. Our findings suggest synaptic adhesion as a key function in compulsive behaviors across three species, and demonstrate how combining targeted sequencing with functional annotations can identify potentially causative variants in both coding and noncoding regions, even when genomic data is limited. ER -