A hybrid AI approach combining deep learning with physics simulations designed stable peptides targeting the IL-23 receptor for autoimmune disease.
AI + physicsHybrid approach combining deep learning peptide generation with molecular dynamics validation
What the researchers found
LSTM-generated peptides with GRU-predicted anti-inflammatory activity showed stable IL-23R binding in molecular dynamics simulations.
Why it matters
Peptide drugs targeting IL-23R could be cheaper and more accessible than antibody therapies for common autoimmune diseases.
The numbers in context
Peptide P4: IC50 = 2 uM; LSTM for generation, GRU for classification, MD for validation; confirmed IL23R specificity.
How the study worked
Hybrid computational pipeline: LSTM peptide generation, GRU anti-inflammatory classification, and molecular dynamics simulations.
Who was studied
N/A
What this study cannot tell us
Computational study — designed peptides need experimental synthesis and biological testing.
How to read the evidence
Computational design study — novel methodology but no experimental validation of biological activity yet.
When this study was published
Published in 2025, representing the cutting edge of AI-driven peptide therapeutics design.
The bigger picture
AI-driven peptide design could democratize drug development by rapidly generating candidates for challenging therapeutic targets.
Questions still open
- Do the designed peptides actually suppress IL-23-driven inflammation in cells?
- How do they compare to approved anti-IL-23 antibodies in potency?
Common questions
Can AI design new drugs?
What is IL-23 and why target it?
Read the original research
A hybrid protocol for peptide development: integrating deep generative models and physics simulations for biomolecular design targeting IL23R/IL23.
International journal of biological macromolecules, 316(Pt 2), 144652
Citation
Qayyum, Naila; Seo, Hana; Khan, Noman; Manan, Abdul; Ramachandran, Rajath; Haseeb, Muhammad; Kim, Eunha; Choi, Sangdun. (2025). A hybrid protocol for peptide development: integrating deep generative models and physics simulations for biomolecular design targeting IL23R/IL23.. International journal of biological macromolecules, 316(Pt 2), 144652. https://doi.org/10.1016/j.ijbiomac.2025.144652