Predictive, LLC offers computational models and software to predict chemical bioactivity and toxicity
WHAT IT IS: In-silico alternatives to animal testing for acute systemic and topical toxicity
More cost-effective and faster than other methods
WHO DEVELOPED IT: Developed in the Laboratory for Molecular Modeling at the UNC Eshelman School of Pharmacy (Head Professor Alexander Tropsha)
Backed by over nearly 30 years of computational research
Predictive, LLC Software Information:
Founded in 2019 out of Dr. Alex Tropsha's Lab at the UNC Eshelman School of Pharmacy and lead by the team of
Kevin Causey, Alex Tropsha, Eugene Muratov, and Vinicius Alves.
Predictive, LLC’s mission is dedicated to using artificial intelligence and machine learning techniques to understand how pharmaceuticals, chemicals, and personal care products affect people. An area we see that has an unmet need is a software platform that can leverage the existing 6 Pack of toxicology assays database to make predictions about products and the mixture of those products and their potential toxicity. Current safety testing for new compounds entering commerce is expensive and inefficient, relying too heavily on animal testing with questionable relevance to human safety.
In these applications, we provide a predictive platform that makes use of standardized experiments performed in past GLP studies to better inform early safety decision making. This work will improve public health by increasing the throughput of efficiency and confidence in the toxicology assessment. These software packages will have an instant impact on how toxicology studies are analyzed in industry and will not only reduce the number of animals used in testing, but will also provide safer compounds as in vitro assays become more predict to actual human health outcomes.