World Database Developed for Nervous System Repair

“There are just too many studies for investigators to keep up with, so we need ways to allow them to search the literature much more efficiently and find information relevant to spinal cord injury. That’s what we propose to do.”

Awarded by the National Institute of Neurological Disorders and Stroke (NINDS), the four-year grant solidifies UM’s growing reputation as the go-to institution for the development of chemoinformatics tools and ontologies that allow massive and diverse data sets to be integrated, queried, interpreted, and studied across multiple disciplines. In addition to Dr. Lemmon, the grant’s coprincipal investigators are UM’s John L. Bixby, PhD, professor of molecular and cellular pharmacology and neurological surgery and, Stephan Schürer, PhD, research assistant professor of molecular and cellular pharmacology, and Ubbo Visser, PhD, associate professor of computer science.

Tentatively called RegenBase for Regeneration Database, the proposed knowledge-based system will integrate and build on the BioAssay Ontology that Drs. Schürer, Lemmon, and their team of UM programmers and computer scientists developed with a US stimulus grant to enable chemists and biologists on the hunt for new therapeutic agents to rapidly search repositories of thousands of research projects on hundreds of thousands of small-molecule compounds.

Until the BioAssay Ontology was released into the public domain in 2011, the value of such chemical compound repositories to the drug discovery process was limited because researchers could not easily search, compare or query their huge data sets, nor integrate them with other data sources.

RegenBase is designed to remove the same hurdles scientists find while looking through the diverse data generated by neuroscientists and cell biologists studying nervous system injury, disease, and cell motility, all with the anticipation of regenerating nerve function and promoting nerve protection. Currently, their efforts are slowed by the complexity of gathering, analyzing, and displaying data from thousands of different experiments conducted on nervous system injury and interpreting them based on knowledge from other fields, such as genomics, cell biology, cancer, drug discovery, and immunology.

To resolve that issue, the UM team plans to link RegenBase with the BioAssay Ontology, as well as other biomedical domain-level ontologies, which will allow scientists anywhere in the world to link data and findings from studies on gene expression in nervous system injury and disease to data and knowledge from other domains that emphasize molecular targets and small molecules that perturb their function, ultimately speeding the development of novel therapeutics.

“One of the goals is to allow us to find compounds that are active on specific genes or proteins and then use those in new experiments on regeneration and protection,” Dr. Lemmon explained. “For example, in our experiments, we might find a class of genes whose expression changed dramatically after injury. By using RegenBase, we may find these same genes were implicated in neuroprotection, so we could search for compounds that inhibit the activity of the proteins that are coded for by these genes, and test to see if they prevent cells from dying, or improve regeneration. RegenBase will help us open that door.”

Underscoring UM’s growing prominence in chemoinformatics, the BioAssay Ontology project, which is ongoing, was recently leveraged into a new NIH-funded collaboration with scientists at Harvard University (Cambridge, MA, USA), the Broad Institute (Cambridge, MA, USA), Columbia University (New York, NY,USA), Yale University (New Haven, CT, USA), and the University of Arizona (Tuscon, USA), to integrate more molecular and biochemical data from the Library of Integrated Network-based Cellular Signatures (LINCS) program.

Related Links:
University of Miami