Using AI in medical research with Carterra’s LSA platform at Absci

0:00

from Siri on your phone to bots like chat GPT artificial intelligence is playing a role in a growing number of areas of our lives but some medical researchers have higher hopes for the technology and one Vancouver lab is using it to develop drugs that they hope will lead to some crucial breakthroughs including Cancer Treatments Fox wolves Karli Olson got the chance to visit absci. she joins us live in the studio with that story.

0:24

Karli Olson: that's right Pete and Riel the method of developing medications through artificial intelligence only came onto the scene a few years ago but since then the results have spoken for themselves and those working in the lab all the way up to top cancer doctors can't deny its potential for the future of medicine take a look.

O: 43

Sean McClain: we really see this technology as revolutionizing the way that uh drugs are created for patients

0:50

Karli Olson: we suit up to explore the inner workings of absci lab in Vancouver that Shawn McClain originally started for the purpose of using e coli to produce new antibody based drugs.

1:00

Sean McClain: The reason why we wanted to do this was to ultimately reduce the biomanufacturing cost and reduce the time it took to get these Cutting Edge therapies to patients.

1:11

Karli Olson: but in recent years artificial intelligence introduced a way to speed up that process and absci jumped at the chance.

1:17

Sean McClain: I am sure that uh you know you and others have heard of chat GPT these large language models that are essentially trained on the whole internet and are able to do uh you know really interesting uh new novel tasks and we're essentially taking that same concept uh and applying it to drugs.

1:38

Karli Olson: the lab feeds millions of data points into artificial intelligence models which the models then use to produce new drugs for certain diseases.

1:45

SM: Essentially think of it instead of searching for a needle in the hay stack you're actually creating the needle in our case the drug of Interest this takes years off of the guessing game of the normal development process normally it takes 5 and A2 years to get a drug into clinical trials.

2:02

we're able to show with our with the lead drug that we have for IBD inflammatory bowel disease that we can get a drug uh into the clinic or into clinical trials within 24 months time is money and all that Sav time lowers the cost of getting a drug to clinical trials from between 50 and $100 million to between 13 and 15 million I think as a scientist who's been doing both lab-based and clinical research for 40 years I thought that was over the top.

2:35

Dr. Bernard Fox is the harder family chair for cancer research and the chief of molecular and tumor Immunology at the Providence Cancer Institute these are actually human cancer cells the hype around AI drug development made him feel skeptical until he read a research paper last summer detailing findings from a lab that used AI to discover how tea cells can recognize not just one but multiple forms of cancer in a patient they had used AI to predict that of how the structure of interactions are between a t- cell and what they see so that really rocked my world in terms of thinking that's incredible how that's

3:10

something we never would have really envisioned since then he's become an advocate for the method hopeful for what the future could bring a lot of the drug development that's out there is guessing people are trying to figure out um what is the right next thing they need to do now you don't have to guess so much now you can go ahead and design it play with it see how it interacts and move it forward the speed and effectiveness of the method has prompted manufacturers to reach out to absci for help developing drugs for things like

3:40

skin diseases and irritable bowel syndrome and most recently Absci partnered with AstraZeneca to develop a drug to treat cancer McClain believes within the next decade it will be used to narrow down the right drug for the right patient faster and potentially save lives you're going to be able to develop drugs cheap enough for an individual and you're going to be able to ultimately I believe in the next 8 to 10 years take a patient sample find out the target that's relevant for that disease and then design a drug that treats that disease for that given patient and that includes cancer patients and move them into patients the right patient um at the right time right not to wait until they're dancers more progressed because they failed other drugs that aren't going to work give them the right drug the first time the American Association for cancer research just wrapped up its annual meeting in San Diego today and multiple drug development companies presented their findings in the search for new cancer drugs using artificial intelligence programs reporting live in the studio Kari Olson Fox2 Oregon.