Jonathan D. Wren, Ph.D.

Biography

Scientific progress is predicated on the ability to observe, hypothesize and test, but we live in an age of data with no shortage of reported observations in many different formats. My research is in Bioinformatics and I am interested in developing methods to sift through the massive amounts of public data available (literature, gene expression, protein-protein interaction, genomic annotations, etc.) to identify patterns and implications of interest in biomedical research. For example, in 2007, approximately one-third of human genes were of unknown function, making it difficult to form hypotheses to test without any prior knowledge to build upon. So I developed a novel method to predict gene function and phenotype using transcriptional correlation networks from public microarray data in GEO in combination with literature-mining. Wet-lab tests have shown it is very accurate at phenotype prediction (45/54 or 85%), and has led to a potential therapeutic treatment for gliomas. My work in computational inference also extends to analysis of literature (>27 million scientific papers published to date) and genomic analysis (>6,000 annotations of different types for the human genome). In short, I am interested in developing technological solutions to bridge the gap between our own limited observational power as individual researchers and the massive number of observations that continue to be collected at an exponential pace.

Email

Jonathan-Wren@OMRF.org

Publications

Research Interests:

• Bioinformatics
• Data integration
• Literature mining
• Transcriptional networks
• Artificial Intelligence
• Emergent properties of complex networks