Using molecular dynamics simulations to prioritize and understand AI-generated cell penetrating peptides.
Duy Phuoc Tran, Seiichi Tada, Akiko Yumoto, Akio Kitao, Yoshihiro Ito, Takanori Uzawa, Koji Tsuda
Author Information
Duy Phuoc Tran: School of Life Sciences and Technology, Tokyo Institute of Technology, 2-12-1, Ookayama, Meguro-ku, Tokyo, 152-8550, Japan.
Seiichi Tada: Emergent Bioengineering Materials Research Team, RIKEN Center for Emergent Matter Science, 2-1 Hirosawa, Wako, Saitama, 351-0198, Japan.
Akiko Yumoto: Emergent Bioengineering Materials Research Team, RIKEN Center for Emergent Matter Science, 2-1 Hirosawa, Wako, Saitama, 351-0198, Japan.
Akio Kitao: School of Life Sciences and Technology, Tokyo Institute of Technology, 2-12-1, Ookayama, Meguro-ku, Tokyo, 152-8550, Japan.
Yoshihiro Ito: Emergent Bioengineering Materials Research Team, RIKEN Center for Emergent Matter Science, 2-1 Hirosawa, Wako, Saitama, 351-0198, Japan.
Takanori Uzawa: Emergent Bioengineering Materials Research Team, RIKEN Center for Emergent Matter Science, 2-1 Hirosawa, Wako, Saitama, 351-0198, Japan.
Koji Tsuda: Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5 Kashiwa-no-ha, Kashiwa, Chiba, 277-8561, Japan. tsuda@k.u-tokyo.ac.jp.
Cell-penetrating peptides have important therapeutic applications in drug delivery, but the variety of known cell-penetrating peptides is still limited. With a promise to accelerate peptide development, artificial intelligence (AI) techniques including deep generative models are currently in spotlight. Scientists, however, are often overwhelmed by an excessive number of unannotated sequences generated by AI and find it difficult to obtain insights to prioritize them for experimental validation. To avoid this pitfall, we leverage molecular dynamics (MD) simulations to obtain mechanistic information to prioritize and understand AI-generated peptides. A mechanistic score of permeability is computed from five steered MD simulations starting from different initial structures predicted by homology modelling. To compensate for variability of predicted structures, the score is computed with sample variance penalization so that a peptide with consistent behaviour is highly evaluated. Our computational pipeline involving deep learning, homology modelling, MD simulations and synthesizability assessment generated 24 novel peptide sequences. The top-scoring peptide showed a consistent pattern of conformational change in all simulations regardless of initial structures. As a result of wet-lab-experiments, our peptide showed better permeability and weaker toxicity in comparison to a clinically used peptide, TAT. Our result demonstrates how MD simulations can support de novo peptide design by providing mechanistic information supplementing statistical inference.
References
Nucleic Acids Res. 2004 Jan 1;32(Database issue):D590-2
[PMID: 14681488]
FEMS Microbiol Lett. 2014 Aug;357(1):63-8
[PMID: 24888447]
J Gen Virol. 2002 May;83(Pt 5):1173-1181
[PMID: 11961273]
