The Silicon Trend Tech Bulletin

Icon Collap
...
Home / AI / ML / Alphabet's New AI Lab Plans on Discovering Drugs

Alphabet's New AI Lab Plans on Discovering Drugs

Published Mon, Nov 08 2021 15:28 pm
by The Silicon Trend

GoogleAI
 

 

 

Alphabet's New AI Lab Plans on Discovering Drugs

On Thursday, Google-parent firm Alphabet Inc. officially announced the launch of a new drug discovery London-based company - Isomorphic Labs. The firm will leverage the technologies coming from the DeepMind lab to boost the discovery with an AI approach. However, the CEO and co-founder of DeepMind - Demis Hassabis, made it clear that the launch won't affect the ongoing research works of DeepMind.

 

 

Pandemic Pacing Technologies

In 2020, DeepMind made an innovative leap with AlphaFold 2.0, an AI system to predict the protein 3D structure directly from their amino acid sequence down to atomic precision. Then, as pandemic became the man of the show, the focus shifted to clinicians and scientists working in the labs to invent effective vaccines. During that time, significant private investment was made in drug discovery, molecular cancer, which was around 4.5 times greater than in 2019.

By 2025, we could expect the drug discovery market to hit $71Bn. DeepMind believes that the primary leverage of top-notch computational and AI methods could streamline scientists' work, pacing up the discovery process.

The primary reason behind the business is to develop effective collaboration, and Demis also plans to form an alliance with biomedical and pharmaceutical firms.

 

 

Google Branding
 

 

Drug Discovery Areas Influenced by AI

AI is attractive for discovering drugs because of its swift leveraging and numerous potentials of the computing techs like ML. Demis added, "AI methods will increasingly be used not just for analyzing data, but to also build powerful predictive and generative models of complex biological phenomena." The different areas of drug discovery where AI can be leveraged includes:

• Drug screening - predicts drug's bioactivity and toxicity and assists in classifying and identifying target cells.

• Drug design helps predict a protein's 3D structure like AlphaFold v2 and discover drug activity and interaction with protein.

• Chemical synthesis helps predict the reaction yield, synthetic route design and develop insights into reaction mechanisms.

• Polypharmacology - helps in multitargeting drug molecules and designing the biospecific molecules.

• Drug repurposing - helps in predicting new use cases and identification of therapeutic targets.