AI and Biotech: Developing New Drugs Faster
The way drugs are developed is being rewritten as we speak. Discovering and approving new drugs is a lengthy process that requires years of development and a dismal success rate, but artificial intelligence (AI) could rev up the process in a major way.
This report was first published on 420 Intel.
On 15 May 2023 Google Cloud launched two new AI-powered tools designed to help biotech and pharmaceutical companies speed up the discovery of new drugs and advance medicine: Target and Lead Identification Suite and Multiomics Suite.
Currently, developing a new drug from the idea to the rollout of a finished product takes 12–15 years and can cost over $1 billion, according to a study published in the March 2021 issue of British Journal of Pharmacology. In addition, identifying a biological target needed for a viable drug can take up to 12 months according to NIH, National Center for Biotechnology Information.
The tools could help replace traditional ones: X-ray crystallography and nuclear magnetic resonance (NMR), two main processes that determine protein 3D structures, the biological target needed for the first step in drug development, but with a high rate of failure.
It was announced at the Bio-IT World Conference on May 16-18 at the Hynes Convention Center in Boston, Massachusetts, an event “showcasing the technologies and analytic approaches that solve problems, accelerate science, and drive the future of precision medicine.” The event included speakers from Pfizer Inc., Janssen Pharmaceuticals, and Harvard Medical School.
CNBC reports that Target and Lead Identification Suite is designed to help biotech and pharmaceutical companies predict and understand the structure of proteins, which so happens to make up a fundamental part of drug development. It allows scientists to share and manage molecular data on a protein using Google Cloud’s Analytics Hub to securely exchange data.
Researchers then use data to predict the structure of a protein with AlphaFold2, a machine learning model developed by Google’s subsidiary. AlphaFold2 runs on Google’s Vertex AI, a platform that allows researchers to build and deploy machine learning models faster.
Within “minutes” AlphaFold2 can predict the 3D structure of a protein with high accuracy.
Multiomics Suite, on the other hand, will help researchers ingest, store, analyze and share mass amounts of genomic data. This suite helps researchers embark on genomic data analysis.
CNBC reports that Colossal Biosciences, a biotechnology company that has a goal to use DNA and genetic engineering to “reverse extinction.”
Researchers can sequence DNA much faster than the time it takes to decipher and analyze it. But thanks to technology, genomic data in areas like the genetic variations associated with disease.
The exploration of Investigational New Drugs (INDs) in biotech often overlaps with the cannabis industry, and Google’s new tools could transform business for certain companies.
Shweta Maniar is Google Cloud’s global director of life sciences strategy and solutions. “We’re helping organizations get medicines to the right people faster,” Maniar told CNBC. “I am personally very excited, this is something that myself and the team have been working on for a few years now.”
The first key step of drug development, which is identifying a biological target to focus on and design a treatment around. A biological target is most commonly a protein, a building block of diseases and other life forms. And to find the target, researchers have to identify the structure of a protein, which determines its function in a disease.
“If you can understand the role, the protein structure and role, now you can start developing drugs around that,” Maniar said.
The AI market is projected to reach the trillions if technology keeps up the pace. Google announced OpenAI’s ChatGPT late in 2021, and its generative chatbot Bard in February.