Steroid-responsive genes play a major role in the genetic basis of sexual dimorphism in complex human disease. L. A. Weiss1,2, K. M. Tsang1,2, B. Adviento1,2, K. A. Aldinger3, H. Lee4, K. Kim5, R. J. Schmidt5,6, S. F. Nelson4,7, P. Levitt8,9, D. G. Amaral6,10,11, I. Hertz-Picciotto5,6, C. Ladd-Acosta12, M. D. Fallin13,14, L. A. Croen15, N. Zaitlen2,16 1) Psychiatry, University of California San Francisco, San Francisco, CA; 2) Institute for Human Genetics, University of California San Francisco, San Francisco, CA; 3) Center for Integrative Brain Research, Seattle Children's Research Institute, Seattle, WA; 4) Department of Pathology and Laboratory Medicine, University of California Los Angeles, Los Angeles, CA; 5) Public Health Sciences, University of California Davis Medical School, Davis, CA; 6) MIND Institute, University of California Davis, Davis, CA; 7) Human Genetics, University of California Los Angeles, Los Angeles, CA; 8) Institute for the Developing Mind, Children's Hospital Los Angeles, Los Angeles, CA; 9) Keck School of Medicine, University of Southern California, Los Angeles, CA; 10) on behalf of the Autism Phenome Project team; 11) Psychiatry and Behavioral Sciences, University of California Davis, Davis, CA; 12) Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; 13) Wendy Klag Center for Autism and Developmental Disabilities, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; 14) Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; 15) Autism Research Program, Kaiser Permanente Division of Research, Oakland, CA; 16) Medicine, University of California San Francisco, San Francisco, CA.

   Most complex diseases show sexual dimorphism in prevalence or presentation, sometimes accounting for a large proportion of liability, such as in neurodevelopmental disorders (4-5 fold increased autism prevalence in males) and autoimmune disease (2-3 fold increased rheumatoid arthritis prevalence in females). There have been a multitude of theories to explain this phenomenon, including the contribution of sex chromosomes, the influence of steroid hormones, cultural biases in socialization, and sex-limited phenomena such as pregnancy. In this study, we leverage statistical approaches to examine genetic hypotheses utilizing autism and rheumatoid arthritis (RA) as examples of male-biased and female-biased conditions.
   Our autism dataset consists of imputing from published GWAS datasets as well as uniquely genotyped samples, using family-based and case-control designs to examine 7,462 affected males and 1,503 females. We compared these results with the RA WTCCC dataset (473 affected males, 1,385 females). First, we utilized GCTA to assess SNP-based autosomal heritability (h2g) in males and females separately. Second, we leveraged a local h2g approach (Gusev et al 2014 arXiv) to determine the phenotypic variance explained by estrogen- and androgen-responsive genes. To support both approaches, we also used an FDR-based test for enrichment of association signal. Finally, we use both GCTA and FDR-enrichment to examine the contribution of the X chromosome. Sex and gene set permutations tests were used to assess significance where appropriate.
   For the first time, we show that autosomal SNP-based h2g of several diseases is higher in the lower-prevalence sex (P<0.01). We observe excess h2g for androgen-responsive genes in females across diagnoses (P<0.01). Further, we show an overabundance of high test statistics in androgen-responsive genes in male-prevalent conditions (P=0.05) and in estrogen-responsive genes in female-prevalent conditions (P<0.01). Finally, we quantify the contribution of the X chromosome to heritability by sex.
   In summary, we support the liability threshold model for common variants, describe the contribution of the X chromosome, as well as implicate a specific and major role for hormone-responsive genes in complex disorders. Our results increase understanding of the underlying mechanism of sexually-dimorphic risk and carry important implications for future studies.

You may contact the first author (during and after the meeting) at