Drugmakers use AI for drug discovery

Pharmaceutical giant Bayer has partnered with Google Cloud to use its AI services for a host of operations, including drug discovery and clinical trials.

Bayer has contracted Google Cloud’s Tensor Processing Units (TPUs), chips which allow machine-learning AI programs to process complex data matrices. Using TPUs in conjunction with Google Cloud’s web services, the aspirin maker intends to design a program which can perform quantum chemistry calculations in the search for new drugs.

Other Google products, such as Vertex AI and Med-PaLM 2, will be used to streamline the clinical trials process, including compiling the documentation submitted to regulators like the Food and Drug Administration (FDA).

“[T]hese tools will be able to better decipher large data sets, find correlations between disparate data points, and generate insights that can tangibly be used to advance the research and development lifecycle,” reports Forbes.

Bayer is not the only company using AI technology to help create new drugs.

Insilico, an NVIDIA-backed startup, has used generative AI to discover and design a drug that is now entering Phase II clinical trials. The medication was designed to treat idiopathic pulmonary fibrosis, a respiratory disease which causes lung scarring. Insilico has used generative AI for every step of the drug discovery process, including identifying target molecules, designing drug compounds, and predicting outcomes.

“This first drug candidate that’s going to Phase 2 is a true highlight of our end-to-end approach to bridge biology and chemistry with deep learning,” announced Insilico Medicine CEO Alex Zhavoronkov in June. “This is a significant milestone not only for us, but for everyone in the field of AI-accelerated drug discovery.”

Insilico’s INS018_055 drug will be tested on several hundred people in the US and China — Insilico is headquartered in both countries.

The company has created a platform called Pharma.AI which is doing the same for at least 12 other drugs. One of them, a COVID-19 treatment using viral replication, has entered Phase I.

“When we first presented our results, people just did not believe that generative AI systems could achieve this level of diversity, novelty and accuracy,” said Zhavoronkov. “Now that we have an entire pipeline of promising drug candidates, people are realizing that this actually works.”