Scientific Program | Drug Discovery | Medicinal Chemistry

Protein-binding affinity model expands role of AI in drug discovery

2025-06-09 - 2025-06

Understanding how molecules interact is central to biology: from decoding how living organisms function to uncovering disease mechanisms and developing life-saving drugs. In recent years, models like AlphaFold changed our ability to predict the 3D structure of proteins, offering crucial insights into molecular shape and interaction.

But while AlphaFold could show how molecules fit together, it couldn't measure how strongly they bind—a key factor in understanding all the aforementioned properties. That missing piece is where MIT's new AI model, Boltz-2, comes in.

Boltz-2 breaks new ground by jointly modeling both structure and binding affinity, a critical parameter in small molecule drug discovery. Boltz-2's affinity module was trained on millions of real lab measurements, showing how strongly different molecules bind to proteins. Thanks to this, Boltz-2 can now predict binding strength with unprecedented accuracy across several benchmarks reflecting different stages of real-world drug discovery.

In established benchmarks, Boltz-2's predictions come very close to those produced by full-physics free energy perturbation (a precise computer simulation that predicts how strongly a drug sticks to its target, but that can take up to a day to run one test even on a GPU)—at more than 1,000 times the speed. It's the first deep learning model to deliver that level of precision.

"This performance increase makes Boltz-2 not just a research tool, but a practical engine for real-world drug development," says MIT CSAIL Ph.D. student Gabriele Corso. Corso, alongside fellow student Jeremy Wohlwend and MIT Jameel Clinic researcher Saro Passaro, was a lead researcher of both Boltz-1 and Boltz-2.

"Instead of spending hours simulating the interaction between a single molecule and its target, scientists can now screen vast chemical libraries within the same time frame, enabling early-stage teams to prioritize only the most promising compounds for lab testing."

In early 2023, a team of researchers at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) and the MIT Jameel Clinic for Machine Learning in Health (Jameel Clinic) launched an ambitious experiment: Could a machine learning model not only predict molecular structures, but understand how biomolecules behave—how they interact, and, most critically, how likely they are to bind—a key mechanism in drug discovery.

At the core of many diseases are misregulations in biomolecular functions. This can be mitigated, however, by designing molecules that can bind to the implicated targets. Accurately predicting this binding behavior is one of the biggest challenges in designing effective new drugs.

In 2024, that project became known as Boltz-1, an open-source model designed as a fast, accessible alternative to AlphaFold3, the state-of-the-art method in the field. Since its release, Boltz-1 has been used by thousands of scientists across leading academic labs, biotechs and pharmaceutical companies—making it the most widely used model of its kind in the industry.

Now, the same team, working with biotechnology company Recursion, unveiled the next step: Boltz-2, a major advance in AI-powered molecular modeling.

Boltz-2 improves on Boltz-1 in several key ways. The model has been retrained using a much larger and more diverse dataset, including computer simulations of molecules in motion and synthetic data made from predictions from the earlier version of the model. It also adds a new feature called Boltz-Steering, which helps fine-tune the results by using physics-based cues to make the predicted structures more realistic.

Wrong medicine: US price control push risks global drug discovery

2025-05-15 - 2025-05

US President Donald Trump’s latest executive order seeks to slash prices of prescription drugs by 30-80%. The idea behind the move is to let US consumers “pay the same price (for medicines) as the nation that pays the lowest price anywhere in the world”. The order provides a 30-day window for implementation, but lacks concrete details about how the cost of the price cuts will be shared among the stakeholders. In any case, Trump’s move amounts to a radical policy shift. If implemented in letter and spirit, the order could have serious adverse implications for Big Pharma and half a dozen monopolistic drug distributors — pharmacy benefit managers — in the US market, which corner big margins in proprietary drugs business. The new policy might also negatively impact innovation-driven pharma companies in Europe, Asia, and elsewhere, and to a lesser extent, exporters of generic drugs to the US.

India currently has only a 13% share in the US generic drugs market by value with its annual shipments worth about $10 billion. But it accounts for more than 40% of such products sold in that country, as Indian medicines are among the least expensive. So, the formula to bring US prices to the “most favoured nation” prices might not require Indian drug makers to adjust their prices any further in most cases. To that extent, adverse impact on their export business will be muted. What, however, causes concern is that tight price controls on patented drugs in the world’s largest market could spell a death knell for new drug research and discovery. For the world’s largest economy, which ranks seventh in per capita income, to expect the prices of prescription medicines to be at par with the lowest in any nation is unreasonable.

Trump’s approach here, as with his policies related to import tariffs, and climate, is odd and self-defeating. As such, the pipeline of new chemical entities/biologicals, with breakthrough therapeutic value and commercial blockbuster status, has dried up since the dawn of the millennium. Drug development time-lines haven’t seen any major reduction either. Attrition rate among drug candidates remains frustratingly high, causing R&D costs to spiral out of control. Launch of new patented drugs are not keeping pace with patent expiration. The ordinary American has low-cost options for medicines — nearly 90% of the US pharma market by volume is constituted by generic products, while in value terms these account for barely 15%. Those who buy patented products are the rich, who can afford to pay, and fund mankind’s quest for vastly improved healthcare systems. Trump wants cost of innovation to be shared “equitably” by all stakeholders. In effect, as proprietary medicines fail to recoup costs, a higher burden of healthcare goals would fall on taxpayers.

For the Indian drug industry, especially the creamy layer of it, the development must be a wake-up call. They have long benefitted from India’s intellectual property rights policies, which militate against ever-greening of patents, and deny extended “data exclusivity” to patent holders. As Big Pharma is likely to turn more risk-averse, Indian companies can’t rest on their past laurels. They ought to seize the opportunity to augment their R&D endeavour. Indian firms can be among the early movers in the use of artificial intelligence for design, testing, and efficacy analysis of drug candidates, and novel in vitro techniques to develop personalised treatment options.

2nd Webinar on Drug Discovery

info@globalmeetx.net

ravendev55@gmail.com

+44-7723-860457

Protein-binding affinity model expands role of AI in drug discovery

Wrong medicine: US price control push risks global drug discovery