The potential of AI to overcome fundamental drug discovery stumbling blocks and help develop effective medicines faster has been recognised and harnessed across the pharmaceutical industry.
This article looks at some of the applications of AI in drug discovery and how, by building their own internal AI teams or partnering with experts, pharma companies are lowering costs and making processes more efficient.
Applications of AI in drug development
Target identification
The drug development process typically starts with the identification of a potential target for a drug. The target could be an enzyme or receptor in the body, or an organism that causes disease.
Traditional target identification is a time-consuming process that can take many years, but AI’s ability to analyze large datasets and intricate biological networks means it is now playing a vital role in drug target identification.
Virtual screening and optimization of compounds
Once the target is identified, researchers then search for a compound that can interact with it and modulate its activity.
Advanced AI algorithms can streamline the identification of new drug candidates by rapidly analyzing vast databases of existing compounds and biological data. By developing predictive models, AI can pinpoint the compounds with the highest probability of binding to the target protein.
Once those compounds that work best for a specific biological target have been identified, they go through rigorous preclinical testing. AI can be used to predict the toxicity of a compound, and any probable side effects. This saves researchers significant amounts of time, not just on the analysis but also in reducing the number of compounds that they then put forward for efficacy and safety testing in the lab.
Drug repurposing
Drug repurposing seeks to discover new applications for an existing drug that were not previously referenced, are not currently prescribed, and have not been investigated.
It builds upon previous research and development efforts, meaning R&D costs are considerably lower and, while clinical trials are still required to demonstrate the drug’s effectiveness in the new area, the notoriously expensive and lengthy phase 1 trials can be avoided as safety in humans has already been established.
There will also be plenty of previously generated data available, including comprehensive information on the drug’s pharmacology, dose, possible toxicity, and formulation, so therapies can be ready for phase 2 clinical trials quickly.
AI can help make drug repurposing a speedier and more streamlined process, with the capacity to quickly and efficiently sweep a search space of drug candidates far beyond human capability.
AI can give insight into the biomolecules involved in disease causation and treatment, generate better matches between repurposed drugs and target proteins, and identify any potential adverse side effects.
Benefits of AI use in drug discovery
As well as speeding up every aspect of drug development, simplifying processes, and vastly improving accuracy, AI helps reduce the need for phase 1 clinical trials. Given clinical trials take many years and millions of dollars, this is enormously beneficial financially for pharma companies.
Add to that the potential for swiftly and effectively repurposing drugs to meet pressing medical needs and you can quickly see just how transformative AI is already proving itself to be in this industry.
Real success stories
AI is already revolutionizing pharma, here are just some examples:
Sanofi partnered with an AI startup in 2018 to build an AI document processing solution to automate medical literature reviews using natural language processing (NLP). The resulting solution reduced review time from 13 minutes per paper to just one second, while also increasing the accuracy.
In 2019, Pfizer announced a partnership with Concreto HealthAI, a precision oncology company to identify new and more precise treatment options and speed up their redefined study designs.
Janssen Pharmaceutica, part of Johnson & Johnson, collaborated with a French AI start-up in 2019 to develop an AI-powered drug design system that enables fast identification of molecules to meet the desired criteria in each research project.
Bayer partnered with Exscientia, a UK-based AI-driven drug discovery company, between 2020 and 2022. The collaboration identified new difficult-to-reach cardiovascular and oncology targets.
Excitingly, the first fully AI-generated drug entered phase 2 clinical trials earlier this year, after successful phase 1 results in January.
Insilico Medicine is a Hong Kong-based AI-powered drug discovery company, and their drug, INS018_055, is being trialed for the treatment of a rare, progressive chronic lung disease called idiopathic pulmonary fibrosis.
AI played a crucial role in discovering the specific protein target the drug binds to, and the timeline of the drug development process to clinical trial was just one year, a striking contrast to the usual average of five years.
Download your free guide on The Impact of AI in Life Sciences by clicking the link below!
