Peptides are highly attractive candidates as therapeutic drugs and adjuvants. However, discovering peptide sequences with a desired bioactivity remains challenging, as even small sequence variations can drastically alter physicochemical properties across multiple length scales. Here, I present how integrating experimentation, machine learning, and molecular simulations accelerates peptide discovery.
Using viral infectivity enhancement as a model system, we synthesized and screened 160 peptide derivatives to identify structure-activity relationships. Surprisingly, bioactivity does not originate from fibrils alone, but from μm-sized β-sheet-rich aggregates that interact with cells and viruses.[1,2] Molecular dynamics simulations revealed that hydrophobicity, both at molecular and nanoscale levels, controls aggregation behavior.[3,4] These insights enabled us to engineer amyloid-type peptides for spatially controlled cell-attachment and extraordinarily high degradability.[4,5]
Building on these mechanistic insights from individually designed peptides, we developed scalable discovery workflows for functional peptides. A machine learning model trained on the infectivity-enhancing library predicted new bioactive sequences in chemical space entirely distinct from the training set.[6] As a complementary discovery strategy, we employed high-throughput combinatorial libraries using fast-flow peptide synthesis.[7] By synthesizing over 40,000 peptide sequences and screening them for liquid–liquid phase separation using tandem mass spectrometry, we identified phase-separating peptides with 86% accuracy. Strikingly, this library-based approach reveals behaviors inaccessible in homogeneous solutions such as insights into how aggregation-prone sequences are stabilized within condensed phases.
Overall, this lecture will demonstrate how combining rational design, computational prediction, and high-throughput methods rapidly expands the peptide toolkit for functional biomaterials and therapeutics.
References:
[1] K. Kaygisiz, L. Rauch-Wirth, A. Dutta, X. Yu, Y. Nagata, T. Bereau, J. Münch, C. V. Synatschke, T. Weil, Nat. Commun. 2023, 14, 5121.
[2] J. La Roche, L. Rauch‐Wirth, L. Zimmerman, F. Zech, J. Münch, C. Read, K. Kaygisiz, J. Pept. Sci. 2025, 31.
[3] A. Iscen, K. Kaygisiz, C. V. Synatschke, T. Weil, K. Kremer, Biomacromolecules 2024, 25, 3063.
[4] K. Kaygisiz, L. Rauch‐Wirth, A. Iscen, J. Hartenfels, K. Kremer, J. Münch, C. V. Synatschke, T. Weil, Adv. Healthcare Mater. 2024, 13, 2301364.
[5] K. Kaygisiz, A. M. Ender, J. Gačanin, L. A. Kaczmarek, D. A. Koutsouras, A. N. Nalakath, P. Winterwerber, F. J. Mayer, H. Räder, T. Marszalek, P. W. M. Blom, C. V. Synatschke, T. Weil, Macromol. Biosci. 2023, 23, 2200294.
[6] K. Kaygisiz, A. Dutta, L. Rauch-Wirth, C. V. Synatschke, J. Münch, T. Bereau, T. Weil, Biomater. Sci. 2023, 11, 5251.
[7] K. Kaygisiz, B. L. Pentelute, ChemRxiv 2026, 12, 06.
