Network Analysis of Genomic Variants to Identify Causative Variants in Complex Disease
My primary research interest was to understand how genetic variants combine to generate complex diseases. Complex diseases cannot be explained by a single genetic factor, but are instead caused by multiple factors and heterogeneous modifications to regulatory systems. My primary focus was on Autism and related mental disorders. Autism is a complex developmental disorder that shows an estimated 60-80% genetic heritability, but only 10-20% of known cases have isolated genetic contributing factors. My research was focused on developing novel network-enrichment methodologies that take advantage of the network and regulatory structure present in biology to identify the systemic disruptions that occur, as opposed to individual genetic mutations.
To further this goal, I developed a network-enrichment method, implemented as a suite of programs that performs a full network-based gene-set enrichment pipeline, including importing networks, SNP, and p-value data, creating permuted networks, generating enriched subnetworks from native and/or permuted data, and calculating complex subnetwork statistics. It used a novel network growth algorithm that I developed, which attempts to maximize the significance (calculated as negative log odds) for all SNPs in the subnetwork, while minimizing the subnetwork size. This algorithm was designed to be able to filter out random noise, and provide a higher signal-to-noise ratio for determining potential causative genes or subnetworks in complex diseases.