The new strategy combines predictive and generative AI to design new chemical entities with specific biological effects. Credit: IRB Barcelona
The classical drug discovery paradigm begins with a known molecular target: a protein whose modulation is expected to reverse the course of a disease. However, in many pathologies, such a target does not always exist or is not sufficiently characterized.
Now, the Structural Bioinformatics and Network Biology Lab at IRB Barcelona, led by Dr. Patrick Aloy, proposes a new strategy to design molecules based not on a specific protein, but on the effect they are intended to induce in cells.
In this approach, known as phenotypic discovery, the starting point is not a specific molecular target but an observable response in the cell—for example, a molecule acting on a specific cell type and not on others.
To test the methodology, the team used various cell models, including pancreatic cancer-derived lines and control cells.
“For the first time, we have designed new chemical entities using artificial intelligence based on the biological effect we wanted to achieve, and we have experimentally demonstrated that they work on specific cells,” explains Dr. Patrick Aloy, an ICREA researcher at IRB Barcelona.
Pushing the boundaries of screening
To train the system, the researchers first generated their own database by testing more than 11,000 chemical compounds across eight different cell models: six pancreatic cancer lines and two controls. Using these data, they created predictive models based on the bioactivity information of each molecule in the cells, which proved to be much more accurate than methods based solely on chemical similarity between compounds.
They then integrated these models into a generative AI and machine learning system capable of proposing new candidate molecules. The goal was to design new molecules under a dual criterion: that they be active against a specific cell type while having a lesser effect on control cells or other cellular profiles.
Experimental validation: from computer to laboratory
The team experimentally evaluated many of the AI-designed molecules, and several matched the function they were designed for: acting selectively on certain cell models while having a lesser effect on others.
The AI-designed molecules not only demonstrated superior activity compared with those obtained through conventional screening strategies, but many also turned out to be structurally innovative and distinct from known chemical compounds.
Although this is still an early stage of compound discovery, the methodology opens new possibilities for identifying candidate molecules in a faster and more targeted manner, especially in contexts where there is no clear therapeutic target.
Publication details
Gema Rojas-Granado et al, Phenotypic AI-based design of cell-specific small molecule cytotoxics, Communications Chemistry (2026). DOI: 10.1038/s42004-026-02071-x
Provided by
Institute for Research in Biomedicine (IRB Barcelona)
Who’s behind this story?
Citation:
AI-generated compounds hit specific cell types and outperform conventional screening (2026, June 6)
retrieved 6 June 2026
from https://phys.org/news/2026-06-ai-generated-compounds-specific-cell.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.