โMy whole career has been about finding alternatives to animal testingโ
2022๋ 11์ 11์ผ
์ ์: Ann-Marie Roche

AI can learn from past animal studies and generate animal data, reducing the need to conduct new animal studies, says FDA toxicology researcher
Caption: Dr Weida Tong, a lead researcher at the FDAโs National Center for Toxicological Research and Director of its Bioinformatics and Biostatistics division, has created a virtual lab rat as part of his quest to find data-driven methods to test human toxicity. (App image ยฉ istock.com)
Artificial Intelligence can learn from past animal studies to generate new animal data, reducing the need to conduct new animal studies, explains Drย Weida Tongย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ.
That reality is at the heart of his research.
โThe revolution has begun, and over the next few years, weโll see tremendous progress,โ Weida says.
In fact, he already has a virtual lab rat to prove it. But what doesย Americaโs Got Talentย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย have to do with it?
Big data + powerful machines = unprecedented results
โMy field is, by default, trying to replace animal testing,โ Weida explains. โAlternative methodologies such asย in silicoย are simply a new way to talk about the existing paradigm. How do we make sure a chemical is safe for humans? This is what weโve been continually working on.
โBut now the machines are powerful enough to deal with all these tremendous resources of data, such asย PharmaPendium. We are now getting unprecedented results and these results are being evaluated and considered by regulatory agencies.โ
As Director of theย Division of Bioinformatics and Biostatisticsย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย at theย National Center for Toxicological Research (NCTR)ย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย โ the only dedicated research facility for the US Food & Drug Administration (FDA) โWeida obviously has a very serious job. However, with twinkling eyes and an email sign-off that says, โBe nice, play fair, and hire someone smarter than you!โ itโs safe to say he doesnโt take himself overly seriously.
With over 300 peer-reviewed papers and book chapters, Weida has become a champion of data-driven alternatives to animal testing โ making him the perfect guest for Elsevierโsย Successful Alternatives to Animal Testingย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย webinar series.
Webinar series:ย Successful alternatives to animal testing
Elsevier Life Science has created a free webinar series on animal testing:
Watch Dr Weida Tongโs webinar now: Successful AI Alternatives to Animal Testingย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ
Watch Dr Gail Van Normal's webinar now: Successful Alternatives to Animal Testingย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ
Register for the November 16 webinar: Successful Alternatives to Animal Testing: Using Virtual Control Groupsย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ
A short history of ascertaining toxicity
โThis whole field is evolving very rapidly and along the way, the name has changed many times while still basically covering the same things,โ Weida explains. โSo this causes a lot of confusion.โ
Born and raised in Shanghai, Weida earned his PhD there in polymer chemistry. โThis subject really has nothing to do with what I do now,โ he says. โBut when I moved to the US to pursue my career in 1991, I found a job working with so-called computational chemistry, which at that time theorized if we know a chemical structure and how itโs joined, weโll be able to figure out its toxicological profile and perhaps what the biological activity might look like.
โThis was a very big deal at the time because pharmaceutical companies thought it was going to be the magic bullet and weโd be able to foresee new drug candidates. But of course, it was not true. And that happens a lot in this field.โ
For instance, the concept of machine learning was soon brought in as a stand-in for a big chunk of computational chemistry. โBut that also didnโt last because the methods were not yet sophisticated enough. So thatโs why we are talking about AI today,โ Weida says, chuckling.
โAround 1994, a new director at NCTR had come in and wondered: โWe do all these toxicology studies we publish the papers, and then it all disappears into a mass of literature? Why canโt we just go back and use all this data to improve our knowledge and develop new hypotheses?โ So in 1996, I was brought in to create a toxicological knowledge base so we would be able to capture all this generated knowledge and then use it to guide future research.โ
Weidaโs first project was theย Endocrine Disruptor Knowledge Baseย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ, which was very much ahead of its time. The bookย Our Stolen Futureย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ, by Theo Colborn, had just been released and basically collected all the evidence and confirmed what was already written in the bookย Silent Springย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย from 30 years earlier โ namely that environmental pollutants were impacting human health and fertility. The book triggered the writing of a new US law that would ensure any compounds put into food or drinking water did not cause any toxicity or endocrine disruption.
It also sparked theย Environmental Protection Agency (EPA)ย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย to start looking into how they could streamline the testing of over 50,000 known chemicals, which at the time cost $25,000 per chemical. โSo I began developing these models, which we would now call AI,โ Weida recalls. โAt that time and still today, these are calledย Quantitative Structure-Activity Relationships (QSARs)ย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ, which could determine which chemicals were most likely going to disrupt the endocrine system based on their structure.โ
It proved a big job โ and remains a work in progress.
โWeโre still pretty much in the same boat and the EPA is still trying to develop what they call a data-driven scientific approach to address this,โ Weida says. โAt least now, the techniques we are using are more sophisticated.โ
Meanwhile, the proposed law remains unratified.
The toxicogenomic revolution
In the meantime, NCTR had moved to another area called toxicogenomics โ a field made possible by the completion of the Human Genome Project in 2003. โThis revolutionized our entire field, which was now being called bioinformatics โ an existing term โ but then it became a more overarching concept to include all computational techniques dealing with biological information from this new data tsunami,โ Weida explains.
โOne of the resulting ongoing projects from that time is theย MicroArray/Sequencing Quality Control (MAQC/SEQC)ย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย project whereby we used so-called microarrays to test for tens of thousands of gene expressions in a single experiment,โ says Weida.
As a result, the FDA took a close look at the use of genomic technologies in assessing drug and food safety and clinical applications.
โIโm also very proud of theย Liver Toxicity Knowledge Base (LTKB)ย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ,โ Weida adds. โBasically, liver damage is a top reason โ number 1 or number 2, depending on who you talk to โ why a drug is withdrawn from the market. And with the cost of bringing a drug to market about $2.6 billion, you really want to flag these drugs and withdraw them at an early stage.
โOur model aims at asking the right questions during the review process to help determine if a drug is safe or not.โ
โAnd now weโve arrived at today, when everyone is talking about AI. But the only difference is that before, we were working with rather small datasets so the results did not make a big wave; but now, with much bigger datasets, we are able to make a lot more sense.โ
What do deepfakes and animal testing have in common?
Weida is a fan ofย Americaโs Got Talent, and one episode caught his attention: the AI companyย Metaphysicย ์ ํญ/์ฐฝ์์ ์ด๊ธฐย became a finalist by resurrecting Elvis using deepfake technology.
Dr Weida Tong was inspired by Metaphysic co-founders Tom Graham and Chris Ume,ย who demonstrated their AI technology on the popular TV showย America's Got Talent. As singer Daniel Emmet performed, the technology made it appear as though the judges โ and Elvis โ were the ones singing.
โWe usually associate this technology withย synthetic mediaย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ, but thereโs also a positive side,โ Weida says with a smile. โLike withย Americaโs Got Talent,ย we would be able to use the same deep generative adversarial network (GAN)-based framework to do synthetic animal testing.โ
In short, Weida and his team proved they couldย derive new animal results from existing animal studiesย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ. This framework, which they call Tox-GAN, resulted in over 87% agreement between their results and the real data. Certainly, these promising results fit nicely into FDAโs larger project to spearhead innovation.
With theย FDA Modernization Actย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ, Congress asked the FDA to study and establish alternative methodologies as part of the regulatory process. As a result, the agency is following the 3Rs principle of โreplace, reduce and refine animal useโ and using a predictive toxicology roadmap. Their Innovative Science and Technology Approaches for New Drugs (ISTANDย ์ ํญ/์ฐฝ์์ ์ด๊ธฐ) pilot program has set out to discover new digital tools that will help streamline drug development โ including AI-driven methods.
โAll these efforts just make sense,โ Weida says. โConventional animal studies are expensive, time-consuming, labor-intensive, and there are also ethical concerns. Above all, these tests are not fail-safe for human safety. But with the FDA Modernization Act, the train has now left the station, and over the next few years, we will see tremendous progress.โ
Free data for future research
What surprises Weida most is how weโre not yet seeing more virtual animal models:
Weโve been doing animal studies for so long, and all this data is stored somewhere. We need to get the machines to find and learn from this data. At NCTR, we already have a virtual rat that has learned from past rat experiments. So we can know if you treat the rat with certain drugs at a certain dose and a certain time, it will produce certain results.
Itโs not yet a complete replacement, but it can supplement some real experiments:
I think this is a real missed opportunity โ using past investments, which would otherwise be lost, to leverage future research. Iโm surprised not more people are doing it!
๊ธฐ์ฌ์

AR
Ann-Marie Roche
Senior Director of Customer Engagement Marketing
Elsevier
Ann-Marie Roche ๋ ์ฝ์ด๋ณด๊ธฐ