Comment: AI-powered drug discovery

The Winter issue of the Drug Discovery World Magazine is accompanied by an exclusive DDW & SLAS2022 supplement ahead of SLAS2022 in Boston, US on 5-9 February. As we all look forward to the expertise at the event, DDW's Megan Thomas outlines the highlights of the first SLAS 2021 AI-Powered Drug Discovery Symposium in London. As SLAS2022 in Boston drew closer and while international industry professionals prepared to meet in person or online to enjoy a programme which highlights life sciences discovery and research and innovations in laboratory technology, I was able to attend the first ever SLAS 2021 AI-Powered Drug Discovery Symposium from November 8-9 at The Francis Crick Institute in London. A taster of what’s to come, I hope. This two-day event gave a detailed overview of the AI tools available today and facilitated insightful discussions about the future of drug discovery collaborations using AI, machine learning and computational science. One of the themes for SLAS2022 is, 'Building AI into the Lab', so it seems fitting to have had this opportunity in this exclusive DDW/SLAS supplement to reflect on the event. The value of AI in drug discovery and development is becoming increasingly evident, which was made clear by Daniel Ting of Singapore National Eye Centre in his keynote address, Artificial Intelligence in Health: Sky is the Limit. Ting spoke of AI as the ‘fourth revolution', an interesting concept which makes comparisons to the immense impact of the previous revolutions: the steam engine, the age of science and mass production, and the rise of digital technology. To regard AI as the fourth is to assume that humanity is on the precipice of great change, so it is therefore critical that drug discovery and development keep up. Crisis and opportunity Ting shared an anecdote which encapsulated this sentiment somewhat poetically: the Chinese word for ‘crisis’ is Wei Ji (危机). Wei means ‘crisis’, while Ji means ‘opportunity’. One only has to look at the impact and scientific milestones leap-frogged during the Covid-19 pandemic to see how crisis can lead to opportunity: not just through the aid of digital technology, AI and machine learning, but in how it gave us the opportunity to ensure this crisis is not repeated. In a Covid context, we might observe that how without AI, machine learning and technology in general, increased waiting times for testing would have led to increased exposure, as well as delays in a vaccine roll-out. A question SLAS2022 has posed in the lead up to the event is, 'How can AI better help researchers to understand new therapeutic targets and identify potential drug candidates more quickly?' After listening to the speakers at the London event, the phrasing of this question is noteworthy. The notion that AI might replace researchers or doctors seems a major cause for public concern when the topic is broached. But this question’s framing helps to debunk that: AI can assist, improve, speed up... but not replace people. This sentiment was echoed across the presentations, particularly in Iktos’ discussions, led by Sree Vadlamudi, Senior Director and Head of Business Development EU at Iktos. When it comes to knowing what investments labs should be considering, it is panel discussions such as that which took place between Daniel Ting, Xiao Liu (University Hospitals Birmingham NHS Foundation Trust) and Edward Chow (National University Singapore), chaired by Dean Ho (National University of Singapore), which make me optimistic for the future. The discussion, The Clinical Validation of AI for Next Generation Medicine, laid bare the steps which need to be taken for a more targeted future in drug discovery and development. I anticipate that – like SLAS’ AI-Powered Drug Discovery Symposium – SLAS2022 will tackle these topics head-on and enable decision-makers to move forward. Looking ahead Again, it is only possible to know what needs to happen now and next once labs have an idea of what the future holds. An open, informed and forward-looking attitude is critical when prediction is a primary variable. The Shape of Things to Come: Predictive Models of Cancer Fate Fueled by Image-’omics, presented by Chris Bakal from the Institute of Cancer Research, was therefore an invaluable resource for attendees looking to better understand what is required going forward. Ethics of AI Observations on cost and time savings, as well as the necessary levels of expertise required to integrate AI into the future of drug discovery and medicine, are big and important questions, which are often the first to be tackled when presented to decision-makers in the industry. This is because the answers form the bedrock of whether or not it is embraced. However, the third session of the day surprised me, as it covered the topic of the ethics of AI. It surprised me not because I don’t believe it important, but rather the fact that in light of its unquestionable importance, it didn’t come to mind until I was presented with the programme. With something as potentially invasive as AI, it is crucial to have these conversations, and I was awed by the depth to which these experts and researchers conveyed the far-reaching tendrils of AI and its potential. Given the diversity of presentations across this two-day event, I can only imagine the possibility, opportunity and expertise at SLAS2022, given its scope and size. The knowledge that will be shared from the top-tier academic institutions and more than 120 life sciences companies, all nearby Kendall Square, known as the 'centre of the nation's biotech industry', has me certain: the future looks bright for the drug discovery and development industry. Volume 23, Issue 1 – Winter 2021/22 | SLAS2022 supplement