AI Business LensTHE BUSINESS OF AI, FOR PEOPLE WHO TEACH IT OR LEARN FROM IT
Medical Xpress · September 13, 2026

AI virtual cell uses protein dynamics to predict personalized breast cancer treatments

MedicineBiologyBiomedical Engineering
THE AI ANGLE
Simulating dynamic cellular drug responses from time-resolved proteomics

Researchers developed ProteinTalks, an AI virtual cell model trained on more than 38 million time-resolved protein measurements to simulate dynamic cellular responses to anticancer drugs over 48 hours. The model outperformed static gene-activity approaches by accurately predicting responses to unseen compounds, discovering synergistic drug combinations for triple-negative breast cancer, and identifying resistance drivers such as AKR1C3. This breakthrough offers biomedical engineering and clinical oncology educators a compelling example of using perturbation proteomics to guide personalized medicine and drug repurposing.

THE TEACHING ANGLE
Students can debate the computational and clinical trade-offs of capturing high-dimensional, time-resolved proteomic trajectories versus relying on conventional static genomic snapshots for predicting therapy responses.

Read the original at medicalxpress.com   Generate teaching or study materials

Instructors get discussion guides, assignments, and mini-cases. Students and readers get a plain summary, class prep, and an exercise. All built from the full article. Three are free with an account.

More in Medicine