Speak to a Protein: An Interactive Multimodal Co-Scientist for Protein Analysis.
- 1Acellera Labs , C/Doctor Trueta 183 , ,
- 2Universitat Pompeu Fabra , , PRBB, C/Doctor Aiguader 88 , ,
- 3Institució Catalana de Recerca i Estudis Avançats (ICREA) , Passeig Lluís Companys 23 , ,
Abstract
Building a working mental model of a protein typically requires weeks of reading, cross-referencing crystal and predicted structures, and inspecting ligand complexes, an effort that is slow, unevenly accessible, and often requires specialized computational skills. We introduce Speak to a Protein, a new capability that turns protein analysis into an interactive, multimodal dialogue with an expert coscientist. The AI system retrieves and synthesizes relevant literature, structures, and ligand data; grounds answers in a live 3D scene; and can highlight, annotate, manipulate, and see the visualization. It also generates and runs code when needed, explaining results in both text and graphics. We demonstrate these capabilities on relevant proteins, posing questions about binding pockets, conformational changes, or structure-activity relationships to test ideas in real time. Speak to a Protein reduces the time from question to evidence, lowers the barrier to advanced structural analysis, and enables hypothesis generation by tightly coupling language, code, and 3D structures. Speak to a Protein is freely accessible at https://open.playmolecule.org.
Related
- Accelerating scientific discovery with Co-Scientist
- An experiment-calibrated in-silico AI co-scientist discovers chemoselective silicon-nitride passivation for area-selective SiO2 atomic layer deposition
- Built in Minutes, Checked by Script: Instructor-Made Interactive Chemistry Study Guides with Generative AI
- MolExplain: An Interactive Tool for Explainable Molecular Property Prediction
- DeNoMAP: An Interactive Web Application for Multi-Criteria Prioritization of De Novo Designed Compounds
BibTeX
@article{navarro2026speak,
title = {{Speak to a Protein: An Interactive Multimodal Co-Scientist for Protein Analysis.}},
author = {Carles Navarro and M. Torrens and Philipp Thölke and Stefan Doerr and Gianni De Fabritiis},
journal = {Journal of chemical information and modeling},
year = {2026},
month = sep,
doi = {10.1021/acs.jcim.6c01738},
url = {https://europepmc.org/article/MED/42912946},
note = {CC BY},
}
RIS
TY - JOUR
TI - Speak to a Protein: An Interactive Multimodal Co-Scientist for Protein Analysis.
AU - Carles Navarro
AU - M. Torrens
AU - Philipp Thölke
AU - Stefan Doerr
AU - Gianni De Fabritiis
PY - 2026
DA - 2026/09/30
JO - Journal of chemical information and modeling
DO - 10.1021/acs.jcim.6c01738
UR - https://europepmc.org/article/MED/42912946
AB - Building a working mental model of a protein typically requires weeks of reading, cross-referencing crystal and predicted structures, and inspecting ligand complexes, an effort that is slow, unevenly accessible, and often requires specialized computational skills. We introduce Speak to a Protein, a new capability that turns protein analysis into an interactive, multimodal dialogue with an expert coscientist. The AI system retrieves and synthesizes relevant literature, structures, and ligand data; grounds answers in a live 3D scene; and can highlight, annotate, manipulate, and see the visualization. It also generates and runs code when needed, explaining results in both text and graphics. We demonstrate these capabilities on relevant proteins, posing questions about binding pockets, conformational changes, or structure-activity relationships to test ideas in real time. Speak to a Protein reduces the time from question to evidence, lowers the barrier to advanced structural analysis, and enables hypothesis generation by tightly coupling language, code, and 3D structures. Speak to a Protein is freely accessible at https://open.playmolecule.org.
SN - 1549-9596
SN - 1549-960X
ER -