LabGenius, an AI-driven scientific research firm, is using Graphcore’s IPUs (Intelligence Processing Units) to accelerate its drug discovery efforts.
LabGenius is now working on developing new cancer and inflammatory disease medicines. For its potentially life-saving work, the company blends AI, lab automation, and synthetic biology.
Until now, the company’s workloads were performed on standard GPUs. LabGenius claims that switching from Cirrascale Cloud Services to Graphcore’s IPUs in cloud instances lowered model training time from one month to around two weeks.
“Previously we used GPUs and it took us about a month to have a functioning model of all the proteins that are out there,” said Dr Katya Putintseva, a Machine Learning Advisor to LabGenius.
“With Graphcore, we reduced the turnaround time to about two weeks, so we can experiment much more rapidly and we can see the results quicker.”
Specifically, LabGenius is using IPUs from Bristol, UK-based Graphcore to train a BERT Transformer model on a large data set of known proteins to predict masked amino acids. This, the company says, enables the model to effectively learn the basic biophysics of proteins.
“[The system] is looking across different features we could change about the molecule — from point mutations of simpler constructs to the overall composition and topology of multi-module proteins,” explained Tom Ashworth, Head of Technology at LabGenius.
“It’s making suggestions about what to design next… to learn about a change in the input and how that maps to a change in the output.”
In today’s world, one in every two people may develop cancer at some point in their lifetime. Current treatments generally cause a great deal of pain, and while survival rates for most cancers are improving, only about half of those diagnosed live for ten years or more.
AI will help in the invention of novel cancer treatments that are less painful and have a higher chance of long-term survival. While LabGenius’ focus today is on finding new cancer medicines, the business emphasises that the ideas can be extended more broadly to develop new remedies for other deadly diseases that haunt humanity.
“Graphcore has changed what we’re able to do, accelerating our model training time from weeks to days,” adds Ashworth.
“For our data scientists, that’s really transformative. They can move much more at the speed they think.”
Want to learn more about AI and how it is transforming the world from industry leaders? Check out http://gaisa.sg/ happening at new Delhi on September 14th and 15th,2022.
Catch all the latest updates on AI : http://nmai.aicra.org