Predicting the Unknown: A CRISTAL CLEARR Solution

Courtesy Photo | In the Combined Logical Estimation Application for Rapid Results (CLEARR) study, the University of Texas at Austin used four organophosphorus chemicals that are similar in structure to nerve agents. (UTA photo)  

FT. BELVOIR, VA, UNITED STATES

01.04.2022

Courtesy Story

Defense Threat Reduction Agency's Chemical and Biological Technologies Department

The Defense Threat Reduction Agency’s (DTRA) Chemical and Biological Technologies Department in its role as the Joint Science and Technology Office (JSTO) for the Chemical and Biological Defense Program is developing predictive toxicology capabilities under the Computational Rapid Identification and Scientific Threat Analysis (CRISTAL) program using the in silico Combined Logical Estimation Application for Rapid Results (CLEARR) tool. This evolving tool was used recently to predict properties of unknown chemicals, making it the first use of the capability to answer real-world questions about chemical threats and the first time such information would be used to guide further study.

Those with nefarious intentions can now theoretically use advanced artificial intelligence/machine learning (AI/ML) technologies to design potential new chemicals and pathogens—even nominally benign ones—and make them far worse.

DTRA-JSTO is preparing warfighter protection from pathogens and chemicals that go well beyond the traditional list of target agents, including those designed using AI/ML technology, which changes the world of threat analysis. Based on structural similarities, a small group of previously unknown chemical compounds might be just as toxic as known chemicals such as nerve agents, if not worse.

CLEARR assessed several novel compounds from researchers at the University of Texas at Austin (UTA) who are working on a medical countermeasures project to “vaccinate” against nerve agents by synthesizing compounds that should be recognized by the immune system to generate antibodies that can degrade nerve agents. These immune system-stimulating compounds, called organophosphorus transition state analogs, are similar in structure to traditional nerve agents.

DTRA-JSTO discussed with the researchers their concerns that the chemicals could be just as toxic as traditional nerve agents, as well as how to mitigate Information Security and Operational Security issues. Part of the path forward was to estimate how toxic the UTA chemicals are by using the CLEARR tool and to see how toxic they are with respect to classical nerve agents. CLEARR will have a fast turn-around time from several hours to a few days and provide estimated toxicity values to compare to traditional nerve agents.

CLEARR is being developed by SRC Inc. and consists in part of numerous databases and computational models, including open source, needed to predict a compound’s chemical and physical properties, environmental characteristics, reactivity, and toxicology values. CLEARR is at its beta-version stage; however, DTRA-JSTO believed CLEARR is mature enough to provide useful information on the UTA chemicals.

With only chemical structure pictures to work from, SRC entered the structural information into CLEARR using its incorporated drawing function. In less than 48 hours, SRC provided over 70 pages of predictive physical, chemical, and toxicological data on the four structures from UTA based entirely on computer modeling analysis. The CLEARR prediction is that the UTA chemicals are two to three orders of magnitude less toxic than traditional nerve agents, so the chemicals the researchers were concerned about are not worrisome when compared to actual nerve agents.

While there is more development to be done and the CLEARR tool predictions need to be validated, this case shows that CRISTAL/CLEARR will be a useful tool in assessing new, emerging, and unknown chemical threats to warfighters to both create treatments for contamination and to protect them from exposure in the field.

POC: Donald T. Cronce, Ph.D., [e-mail]

• Date Taken: 01.04.2022
• Date Posted: 01.04.2022 16:31
• Story ID: 412449
• Location: FT. BELVOIR, VA, US