Artificial Intelligence (IL) and Machine learning (ML) are expected to facilitate the discovery of new drugs and make them cheaper. Are you aware that almost $2.6 billion is spent searching for new medicines and treatments? Do you also know that a chunk of this amount gets wasted because 9 out of 10 therapies fail in various stages between Phase 1 trials and drug regulatory approvals?
So, it’s only natural that people in drug manufacturing will need to look for more cost-effective measures. Leading pharmaceutical companies like Pfizer use IBM Watson based on Machine Learning technology to look for immuno-oncology drugs. Genentech has adopted an AI-driven system to drive the company’s search for cancer cures. The general belief is that AI and ML will lead to faster and cheaper drugs.
How can Artificial Intelligence be adopted for discovering new drugs?
AI technologies have made inroads into the pharmaceutical world since the last decade. Now, more and more biotech firms using an AI-first approach are expected to discover small-molecule drugs, with many of them already undergoing trials.
The time has come for pharma companies to start planning for a future where AI will be used for drug-making regularly. AI-driven apps are diverse and pharmaceutical companies have to decide how AI can add value to their offerings. For this, they need to understand how much of an impact AI can have on the research and development of drugs.
AI isn’t something that can be offered through a single team or a single tool. To derive value from AI, you must change the whole discovery process. You have to make investments in new skills, technologies, and data.
So, how can it help in small-molecule drug discovery? It can provide access to new and improved chemistry, new biology, improved success rates, and quicker discovery processes. AI can eliminate many of the limitations and challenges in traditional research. It can provide more insights to the teams and redefine long-standing workflows.
The shift from traditional software to asset development partnerships has led to significant investments. Do you know that third-party investment in drug discovery has spiked to $5.2 billion by 2021? This amount doesn’t include the money spent by pharma to develop their capabilities.
AI’s impact on traditional drug discovery is yet to be fully felt. However, AI-based capabilities can speed up and improve the steps to lower the costs of performing costly experiments. AI algorithms can revolutionize tasks like molecule testing. This means you only have to do physical testing when needed for validating results.
AI’s adoption in drug discovery is far from complete. It’s not going to happen overnight. While AI-powered innovations are showing great results, pharma companies still have the advantages of expertise, capital, know-how, regulatory expertise, and branding teams. Companies must develop a roadmap that identifies high-value cases aligned with their discovery programs. Continuing with external partnerships is wise for speeding up the adoption of AI technologies in drug discovery.
But, to work with AI players, pharma companies must change their style of working. They have to train decision-makers about how AI-generated recommendations can be reached. Medical scientists have to be well-versed in analytical methods to understand AI algorithms.
Laboratory work, the bedrock of drug discovery, will have to change to take advantage of AI. Experimental work will now have a supporting role, being used only in areas where AI isn’t working as well or mainly for regulatory purposes. Experimentations will be needed to fill in the dataset gaps to make AI methods even more robust.
To conclude, AI offers a lot of technological advancements that can lead to a paradigm shift as far as drug discovery goes. Many of these advances may feel like sea changes, but soon they will become instrumental in speeding up the discovery of new drugs and cures. Those companies acting boldly and following an articulated strategy for AI adoption are likely to emerge as the big winners.
