Science Wednesday: Reduce + Reuse + Recycle = Results!

Each week we write about the science behind environmental protection. Previous Science Wednesdays.

By Nisha S. Sipes

Who knew that we could help make our environment healthier by using new and recycled data!

This month, I have the honor of being presented the Best Postdoctoral Publication Award by the Society of Toxicology (SOT) for my paper “Predictive models of prenatal developmental toxicology from ToxCast high-throughput screening data.” In other words, I studied how new technologies using both new and old data can determine which chemicals are potentially toxic to development.

Along with many other EPA scientists, I have been researching if it’s possible to predict a chemical’s potential toxicity using efficient new technologies in EPA’s ToxCast program. The ToxCast program is running thousands of chemicals through hundreds of different tests in a “high-throughput screening” (HTS) process. If we’re successful, we will be able to better understand how a chemical is toxic to the body and reduce the need for animal testing—all while saving lots of time and money on testing.

My paper focuses on building computer models to predict the toxic effects of chemicals on prenatal development using two sets of data: traditional toxicity data (gathered from 30 years worth of laboratory studies) and ToxCast data (gathered from HTS methods). I compared the two groups of data, and after crunching the numbers we could show that these new HTS methods could predict results from old-school animal testing for developmental toxicity.

It turns out that the ToxCast data can provide new information about which chemicals are toxic to development. We can also use these new technologies to pick out which chemicals are toxic specifically to rat development or rabbit development without animal testing. That level of specificity was wishful thinking just a few years ago!
Hopefully these models, built from reused and recycled traditional toxicity data, will help pave the way for quickly prioritizing which chemicals need a thorough evaluation and will eventually reduce the need for costly and time-consuming animal toxicity testing.

As we get further along in our HTS research, we can use what we’ve learned from this study to better identify target chemicals that may be toxic to humans.

About the author: Nisha Sipes is a post-doctoral fellow for EPA’s National Center for Computational Toxicology. She joined EPA in 2009 and specializes in using computational approaches to understand chemicals’ developmental toxicology.

Editor’s Note: Attending the SOT meeting in San Francisco? Be sure to catch Nisha next week as she presents her paper and the predictive model: March 13, 2012 at 9:37 AM (Pacific time).