Pengyu Hong, Ph.D.
High-Content Neuronal Screening
With estimates for the number of neurons in the human brain falling in the range of billions, and the number of connections they form being orders of magnitude larger, understanding the interactions operating within the nervous system is a Herculean task. For this reason computational methods of analysis are becoming increasingly important. Dr. Hong described his work in which he is developing methods for sorting through massive amounts of imaging data taken from neuronal cultures in order to extract information about how the morphology of these cells changes over time in response to varying conditions.
High-content neuronal screening has recently become a powerful high-throughput methodology for identifying chemical compounds that regulate neuronal morphology. Such compounds are able to change the patterns of neurite outgrowth and the size distributions of cell clusters. Studies examining the effects of treating neurons with these chemicals are expected to be an important approach in examining the nervous system, as well as aiding drug discovery. A typical high-content neuronal screening project generates hundreds of thousands of high-resolution images containing many interconnected neurons whose morphologies are complex. Therefore, it is a great challenge to be able to accurately and automatically analyze these data. In his presentation, Dr. Hong explained how his group is tackling this problem using their own automated method of analysis.
This method has been one of the key innovations to enable high-throughput genetic and drug discovery screening using neuronal cells. It was applied to the data provided by his collaborators around the world. Their computational results have led to successful follow-up studies. For example, Dr. Hong and his collaborators have identified a set of neural outgrowth genes in a genome-wide gene knock-down screen. Several of these genes were then validated using mice or fruit flies as model systems. In a follow-up study, Dr. Hong's collaborators characterized one of these genes, the Drosophila homolog of Phocein, as a regulator of axonal transport, membrane excitability and organization of microtubule networks. This successful finding served as a proof-of-principle for Dr. Hong's approach.
Dr. Hong went on to describe two other successful applications of this method. First, Dr. Hong and his collaborators were able to identify several promising drug candidates in a chemical-compound screen using a Drosophila model of Huntington's disease. Second, the group gained new insights into the impact of three compounds: phenylalanine, phenylpyruvate and phenylacetate on the activity of the myelin basic protein promoter and the production of myelin sheaths. These successes demonstrate that the high-throughput analysis methods employed by Dr. Hong and his colleagues have promising implications for our understanding of the nervous system, as well as for developing therapeutics related to neurological disorders.