Suicidal behavior, a complex phenotype encompassing attempted and completed suicide, has been shown through epidemiological studies to have a heritable component. Two major factors contribute to this heritability. The first is a preexisting psychiatric disorder, such as bipolar disorder or major depression. These disorders are heritable themselves, and individuals with these disorders are at a higher risk of suicide than the general population. The second contributing factor is thought to be independent of psychiatric disorders and more specific to suicide.
In order to investigate this independent factor, our lab previously conducted a genome-wide association study (GWAS) of attempted suicide in bipolar disorder. This GWAS investigated common variation across the genome and implicated a number of genes and regions in attempted suicide. The first goal of this dissertation was to build upon the results of this GWAS by investigating both common and rare genetic variation through next-generation sequencing techniques. We did this by first conducting a whole-exome sequencing study to investigate rare genetic variation within coding regions across the genome. We did not identify any study-wide significant variation through this study, suggesting that coding variation with both moderately large effect size and moderately low frequency is likely not a major contributor to the heritability of suicidal behavior.
We next used a targeted sequencing approach to investigate several candidate genes and regions that had been implicated through our attempted suicide GWAS and the literature. This dissertation outlines four different candidate gene studies investigating variation in two genes (SAT1 and FKBP5), a set of 16 glutamatergic signaling genes, and one candidate region (2p25). Three of these studies were unable to identify any replicable significant variation, suggesting a need for larger sample sizes and more extensive sequencing coverage. However, the 2p25 study identified three variants that were significantly associated with attempted suicide in males. This finding suggests a sex-specific association of this region with suicide and highlights the need for sex-specific analyses within psychiatric genetics.
The second goal of this dissertation was to functionally assess our most promising findings from the sequencing studies. We did this using the CRISPR-Cas9 genome editing system to delete a segment of our implicated 2p25 region within cell lines. We were able to show that deletion of this segment affected gene expression across the genome, indicating our region does have regulatory potential. This functional assessment represents the first time such an approach has been taken in the field of suicide genetics.
Through the work outlined in this dissertation, we have produced sequencing data across the genome and even more extensive data for a number of candidate genes and regions. Additionally, we performed a functional assessment of one of our implicated regions. Functional characterization of GWAS and sequencing results will help us to better understand the basic biology of disease and, through this better understanding, improve treatment for complex disorders like suicidal behavior.