Category: Drug Target Strategies

1018-D - Novel computational method integrating disparate data types for drug candidate MoA profiling

Tuesday, February 6, 2018
5:00 PM - 6:00 PM

Whether a drug is discovered through phenotypic screening or developed as a target-specific compound, it is a difficult and lengthy process to discover its complete in vivo effect in humans. First, target identification must take into account possible polypharmacologic effects, since even drugs deemed “selective” may actually be acting upon multiple protein targets. Second, the distribution of these targets across the tissues of the body further determines how, and where, the drugs act. To combine these two aspects — namely target affinity and tissue expression — we developed a novel method, called “historeceptomics”, to profile a small molecule’s mechanism of action (“histo” for tissues and “recept” for protein targets).

This computational method combines a drug’s polypharmacologic set of protein targets with the tissue expression patterns of those targets. Using a scoring algorithm combining these two types of data, it generates a target-tissue profile for each drug. If a target-tissue pair has a significantly high score, it indicates that both the affinity and the expression in the tissue is high enough that we consider the target-tissue pair to be significant in the drug’s mechanism of action. In our prototype, we combine experimental and predicted affinity results for human proteins with normal human expression data from microarray studies.

This profiling method has several potential uses in drug discovery and development including the prediction of mechanism of action, off-target effects and toxicology, drug repositioning, and potential drug combinations. In a published case study, we profiled clozapine, an antipsychotic drug, and found that it maps to several known and novel receptors, including the histamine H1 receptor in the superior cervical ganglion (SCG). Interestingly, the SCG innervates the salivary gland, which reveals the mechanism of clozapine’s adverse effect of hypersalivation. Thus, this profiling method is able to explain – and predict – this side effect. The profile is also useful in comparing clozapine to other antipsychotics, helping to uncover the causes of drug-specific effects. We believe this novel drug profiling method is a valuable tool for researchers during the drug development process.

Doreen Tivon

GeneCentrix, Inc.
New York, NY

Dr. Felsovalyi has over nine years experience in bioinformatics and has been leading the scientific endeavors at GeneCentrix since its formation in 2015. Dr. Felsovalyi earned a B.Sc. in Biology at MIT. She earned her Masters and Ph.D. in Biochemistry and Biophysics at Columbia University, working in the area of computational structural biology. She performed structural and bioinformatics studies on the cadherin protein family. After a postdoc at NYU School of Medicine, she co-founded GeneCentrix and has been Research Director since then. She was awarded an NIH SBIR Phase I grant, as well as two supplements, that are funding the development of technologies at GeneCentrix.