After matriculating to the University of Pennsylvania's Cell and Molecular Biology graduate group, Dr. Cole defended his dissertation in May 2015, additionally receiving a certificate in Graduate Training in the Medical Sciences (formerly Howard Hughes Med into Grad Scholars).
BS in Genetics, University of Wisconsin-Madison
Dr. Cole joined the Institute for Biomedical Informatics at the University of Pennsylvania as a Postdoctoral Researcher in July 2015. His research focuses on two key projects:
1) Application of cloud computing to biomedical informatics: Cloud computing is the on-demand use of managed computational infrastructure. Cloud computing has revolutionized the technology sector but academia is slow to adopt. As such, cloud computing is an avenue for increasing reproducibility, performance, scalability, efficiency and resiliency in increasingly data-intensive research. Dr. Cole is a certified Solutions Architect with Amazon Web Services and has developed several cloud-based virtual applications for human genetics researchers. Currently, he is leveraging cloud-based NoSQL databases and distributed computing engines such as Apache Hadoop and Apache Spark to achieve fully reproducible annotation of human genetic variants and analysis of aggregate signals in population-scale human genetics datasets, including pathway burden analysis.
2) Human Genome-Phenome Association: Dr. Cole is a computational geneticist working in genome-wide association studies (GWAS). Human GWAS studies have the potential to uncover pathophysiological mechanisms underlying complex disease, however these studies have failed to provide actionable hypotheses about disease processes that have led to novel therapeutics. Additionally, genotype-phenotype associations have the potential to be deeply confounded by unmeasured covariates. Finally, genotype-phenotype associations are not evidence of causality. All of these factors lead to wide-spread irreproducibility among GWAS results. For this reason, Dr. Cole is developing computational solutions to improve reproducibility and interpretability of genotype-phenotype association studies.