Computational prediction of RNA splicing defects and rescue by small-molecule therapeutics

Panorama Medicine, Inc., Philadelphia PA

Investigators

Abstract

PROJECT SUMMARY Panorama has developed a groundbreaking platform that combines big data and genetic sequencing to identify the RNA sequence of splice events, a cause of up to 50% of genetic diseases. These events are believed to be the cause of diseases such as Becker muscular dystrophy, Duchenne muscular dystrophy (DMD), spinal muscular atrophy (SMA), and amyotrophic lateral sclerosis (ALS). The platform, PANOPLY, predicts thousands of potential disease-causing events. Preliminary data has shown the platform to have an 83% prediction accuracy. PANOPLY, unlike previous platforms, predicts for tissue-specific events in 53 tissues. Again, unlike other platforms, PANOPLY provides the likelihood of that event to occur. Panorama combines PANOPLY with a proprietary screening platform to identify small molecules that correct the disease-causing event. In combination, Panorama can use PANOPLY to predict disease-causing events and then identify potential therapeutic solutions, creating a unique, fast-track, drug development mechanism that leaps past discovery to identify potential causes and treatment candidates for pre-clinical trials. We propose this SBIR project to enhance the performance of PANOPLY by tuning the algorithm to better predict more complex disease-causing events and to demonstrate the innovation’s potential by then identifying small-molecule compounds that rescue PANOPLY predicted events. Specific Aim 1: Improve high-throughput predictions from PANOPLY. This aim uses laboratory data to test and improve the prediction algorithms. To do this, we will reproduce predicted events to assess the computational power for more complex situations. We will then tune the algorithm, which uses hundreds of variables, to perform at a higher level for these more computationally demanding circumstances. Milestone: PANOPLY will use the data to tune for selection criteria to achieve a prediction rate of 40-50%. Specific Aim 2: Confirm the ability of compounds to rescue mutant variants identified by PANOPLY. The second aim brings together PANOPLY and the results of Panorama’s target-agnostic compound screening platform to serve as a drug discovery model. To do this we will identify small molecules that rescue the disease-causing event, and then treat the disease-causing event with those identified small molecules. Milestone: one or more compounds will rescue a clinically significant splice event predicted by PANOPLY.