17857
Cross-Disorder CNV Interactome

Saturday, May 17, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
R. Corominas1, G. N. Lin1, X. Yang2,3, D. E. Hill2,3, M. Vidal2,3 and L. M. Iakoucheva1, (1)Department of Psychiatry, University of California San Diego, La Jolla, CA, (2)Center for Cancer Systems Biology (CCSB) and Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA, (3)Department of Genetics, Harvard Medical School, Boston, MA
Background: Neuropsychiatric disorders such as autism, schizophrenia, bipolar disorder and intellectual disability are major burden to society. Our current knowledge of their underlying pathophysiology remains limited. However, a contribution of genetic factors has been clearly demonstrated. It is now firmly established that rare Copy Number Variants (CNVs) play significant role in the risk of psychiatric disorders. Interestingly, many high-risk rare CNVs cross disorder boundaries and are implicated in several psychiatric disorders.   

Objectives:  To advance our understanding of CNV contribution to psychiatric diseases, we are investigating how the genes from high risk CNVs interact on a protein level. We are testing the hypothesis that cross-disorder CNVs have common and unique interacting protein partners by building and analyzing a cross-disorder protein-protein interaction (PPI) network. 

Methods: We have selected 11 high risk CNVs (containing 169 genes) that are firmly implicated in two or more psychiatric disorders for this study. First, a library of 169 open reading frame (ORF) clones corresponding to cross-disorder CNV genes will be assembled from human ORFeome collection and from our in house clone library. For the cases when the clone is not available, we will clone the gene using commercially available total purified human brain RNA. Then, we will experimentally identify interacting partners for these 169 CNV genes using high-throughput yeast-two-hybrid system. The clones will be screened against the whole human ORFeome 9 (169x17,000 genes) to identify new partners, and against themselves (169x169) to identify directly interacting CNV genes. Finally, we will build and analyze the cross-disorder CNV interactome.

Results: We extracted binary physical PPIs reported for these 169 genes from the public databases and from our recent studies of autism and schizophrenia and observed that 33% of these genes do not have any interaction data available; and only 3.5% of them interact with each other. Furthermore, when the genes with PPIs are merged into the CNV nodes, only 3 CNVs become connected via literature curated interactions (3q29, 7q11.23, 22q11.21), and an additional 5 CNV become connected by the PPIs from our recent studies (1q21, 17q12, 15q11.2, 16p13.11 and 16p11.2). In total, only a small proportion of CNV genes (13/169 or 7.7%) are found to be directly interacting at the protein level. This is likely due to the fact that they have not been tested for interactions with other CNV genes. We are filling this knowledge gap by performing the screens described in the methods section above.   

Conclusions: We are building the protein network connecting genes from high risk copy number variants implicated in four psychiatric disorders. This network will identify sets of protein partners that are common and unique to these disorders. This knowledge will help to expand our understanding of the molecular basis of psychiatric disorders.

See more of: Genetics
See more of: Genetics