Michelle Palmer, Ph.D., director of discovery and preclinical research, chemical biology platform, Broad Institute of Harvard and MIT, described the institute’s approach to integrating novel technologies for identifying small molecules that drive translation research and therapeutics.
The chemical biology and novel therapeutics platforms were established at the Broad to discover small molecules that impact biology and medicine and to innovate the process through which probes and drugs are both discovered and developed.
The Broad enterprise, she explained, “is part of a large network, the MLSMR, funded by NIH. Through that network, we collaborate on 25 new projects a year. Projects start with assay development, then progress to hit selection, and prioritization. The hits are then put through a battery of secondary assays to determine which are the most promising for chemical optimization. We work in close collaboration with our chemistry team. We then provide the data and the chemical tool back to the research community through NIH’s probe report.”
Dr. Palmer said that the Broad applies standard HTS assays as well as novel technologies to prioritize hits coming out of screens and then to discover and understand the mechanism of action of a particular compound. “This knowledge informs the medicinal chemistry process and ability to detect, quantify, and understand the effects of the compound in various cell and animal models.”
The Broad applies a variety of integrated technologies to achieve this level of detail. These include next-generation chemistry (diversity-oriented synthesis), small molecule microarrays (SMMs), microfluidic-based chromatin modifying enzyme assays, multiplexed gene-expression measurements, and SILAC proteomics for target ID.
Dr. Palmer believes that automation has had a big impact on the small molecule characterization process, because, “the more you can do to automate a process and improve data consistency, quality, and reproducibility, the easier it is to think about the data. The data-visualization tools that are evolving to allow us to look at large quantities and types of data permit a much more unbiased approach to identifying the mechanism.”