We recently spoke about the Challenge with Kevin Crofton, the Acting Deputy Director for EPA’s National Center for Computational Toxicology. Thank you for joining us. Could you start off by explaining to us the origin of this Challenge?
A classic phrase from the field of toxicology is that “the dose makes the poison.” In other words, most chemicals are not banned but instead regulated to limit our exposure to a level likely to be without deleterious effects. The generic term “Lowest Effect Level” (LEL) is the lowest dose of a chemical that causes significant effects. The LEL provides a conservative estimate of the dose of a chemical that a person could be exposed to on a daily basis without any expectation of appreciable health risk.
Ideally, every chemical to which we are exposed would have a well-defined LEL, but the full range of studies required to estimate this cost millions of dollars per chemical. Only a small fraction of the tens of thousands of chemicals in commerce have been adequately assessed for potential risks to human health and the environment.
Toxicity Testing in the 21st century (Tox21) is working to determine methods to more effectively and efficiently screen chemicals. The ToxCast research project is one of EPA’s major contributions to the Tox21 effort, which is a collaboration between EPA, the National Institutes of Environmental Health Sciences/National Toxicology Program, National Center for Advancing Translational Science and the U.S. Food and Drug Administration.EPA has now released new toxicity data on 1,800 chemicals that are in industrial and consumer products, food additives and drugs. This new toxicity data is generated from screening chemicals in rapid, automated tests called high-throughput screening assays. As part of EPA’s commitment to gather and share its chemical data in open and transparent ways, all ToxCast chemical data is publically available for anyone to access and use through our new web application, the Chemical Safety for Sustainability Dashboard (CSS). If you are interested in the raw data files, those are also available on the CSS Dashboard.
The EPA is among a select few federal government agencies that is widely embracing – and achieving successful outcomes as result – crowdsourcing and prize competitions as a means to innovate and solve important problems. Was there anything in particular about this Challenge that lent itself to this open problem solving approach?
To make testing faster and more efficient, EPA’s computational toxicology and ToxCast research efforts are harnessing advances in molecular biology, chemistry, toxicology, exposure science and computer science. This research has yielded millions of data points that EPA is sharing with this new release announcement.
However, because the ToxCast chemical data is so massive, the challenge is to make effective use of this huge data volume
Although EPA has developed some solutions for data analyses and use, we want to tap into the broad base of knowledge that solvers exhibit for generating creative solutions. A challenge like this helps us match the scale of our data by leveraging the thousands of solvers out there who could help us with their ideas and insights.
The public data release is the first step in asking external stakeholders to analyze this new data and suggest ways to use it to inform decisions about the safety of chemicals. Stakeholders can also provide valuable insights about improving data accessibility and the navigability of the CSS dashboard.
How will the general public benefit from this challenge?
By reducing EPA’s reliance on expensive animal toxicity tests, ToxCast and Tox21 are working to accelerate the pace of chemical screening. That means our scientists and researchers are developing new tools to evaluate the potential risk of thousands of chemicals at a small cost in a very short amount of time.
Results from this InnoCentive challenge will help inform an algorithm challenge early next year. The future goal is to produce algorithms that quantitatively predict a chemical’s LEL, which will allow better predictions of risks to human health and the environment.
In short, the effort will saves time and taxpayer money while allowing the partner agencies to do a more thorough job of evaluating chemicals for potential risks to human health and the environment.
What are some of the key attributes you’d like to see (or not see) in a winning solution?
Ideally, the winning solution will highlight high-throughput screening assays and chemical descriptors useful in predicting LELs on an administered dose basis and across all classes of endpoints. This is a very difficult task so chemical properties and biological assays alone would be suitable for a submitted solution. Solvers are encouraged to propose additional in vitro assays or in silico models to supplement those that already exist within ToxCast™ and to use unique chemical modeling to solve this challenge.
What guidance would you give to Solvers looking to engage in this challenge?
A great solution will need to pull from a diverse set of expertise, including biology, chemistry and toxicology. Solutions that incorporate integrated, multidisciplinary approaches will be rated higher. Also, we want bold, transformative ideas that will have significant impacts in the field, versus incremental improvements of current methods. However, solutions must still represent feasible approaches that can be reasonably implemented.