Singapore scientists crack disease code with 'game-changing' AI

A new AI model from Singaporean scientists could revolutionise how we understand diseases like cancer and massively speed up drug discovery. This clever tech 'reads' protein pairs, unlocking fresh insights into how our bodies work at a microscopic level.

Game-changing tech for medicine

For years, scientists have grappled with understanding how proteins interact – crucial for nearly all cellular processes. This new "paired protein language model" (PPLM) changes the game. Instead of looking at proteins one by one, it learns from two interacting proteins at the same time. This is a massive shift in how AI is used in biology, helping predict protein-protein interactions (PPIs) with much greater accuracy.

How it works

The research team, led by Professor Zhang Yang at the National University of Singapore (NUS), created PPLM specifically to learn how proteins relate to each other. It does this by 'jointly encoding' paired protein sequences. This captures individual protein features and how they interact with their partners. The model was trained on more than three million protein pairs, giving it serious power to learn these complex patterns.

Stronger, more accurate predictions

The team built on this foundation, developing three specialist tools. These are PPLM-PPI for predicting whether proteins interact, PPLM-Affinity for estimating binding strength, and PPLM-Contact for identifying interaction interfaces. The results are impressive. The model improved interaction prediction accuracy by up to about 17 per cent compared to existing methods. These gains were consistent across different species, proving its versatility.

Notably, it even outshone both sequence-based and structure-based methods in tough scenarios, like understanding how antibodies and antigens interact. The model also identified patterns that perfectly match real-life protein interactions, showing it truly understands the biology.

Professor Zhang explained: “This work highlights the growing role of AI in transforming the life sciences. By moving from single-protein analysis to interaction-aware modelling, the study lays the groundwork for future advances in multi-protein complex prediction, systems-level biology, and AI-guided therapeutic design.”

Future impact on healthcare

Improving the accuracy and scalability of protein interaction modelling means PPLM could support many applications. This includes discovering proteome-scale interactions, identifying drug targets, and speeding up therapeutic development. The NUS team is now working to make the model even better by adding structural and experimental data. They also plan to apply it to more complex biological systems, like how hosts and pathogens interact.

Related Content

MORE: London scientists develop 'smart' drug that wipes out 95 per cent of cancer without toxic side effects — https://trendwiremedia.com/2026/03/07/london-scientists-develop-smart-drug-that-wipes-out-95-per-cent-of-cancer-without-toxic-side-effects/

MORE: Scientists find the 'gatekeeper' protein that keeps your skin waterproof and itch-free — https://trendwiremedia.com/2026/03/23/scientists-find-the-gatekeeper-protein-that-keeps-your-skin-waterproof-and-itch-free/

MORE: Scientists use 1,024 graphics cards to unlock future of drug discovery — https://trendwiremedia.com/2026/04/02/scientists-use-1024-graphics-cards-to-unlock-future-of-drug-discovery/

OFFICIAL SOURCE VERIFICATION: This report is based on official data from National University of Singapore (NUS). Document: NUS scientists devise AI model that “reads” protein pairs, unlocking new insights into disease and drug discovery Source Link: https://news.nus.edu.sg/ai-reads-protein-pairs/