Secondary Structure Bead-Encoded Amphiphilicity Biases Peptide Self-Assembly Prediction in MARTINI Coarse-Grained Simulations.

Sequence-dependent self-assembly of peptides yields ordered supramolecular structures with diverse nanotechnological applications. In the absence of simple design rules linking sequence to supramolecular morphology, coarse-grained molecular dynamics (CG-MD) simulations have become valuable tools for guiding the design of self-assembling peptides. The MARTINI model, despite the lack of explicit hydrogen bonding, can predict self-assembling sequences and structural features by introducing secondary structure-specific beads that adjust backbone polarity. Extended β-sheet encoding is typically used as input for short peptides, based on experimental observations. However, this assumption becomes increasingly unreliable beyond six to ten residues, where folded conformations begin to emerge. In this study, we investigated the effect of different secondary structure encodings on self-assembly simulations of hexapeptides and decapeptides using MARTINI 2.2p. The results confirmed that changes in the secondary structure encoding significantly impact the predicted self-assembly behavior, with AP scores for the same peptide varying by up to one unit─shifting from fully dissolved (AP ≈ 1) to well-aggregated states (AP > 2) in specific cases. This effect arises from alterations in overall peptide amphiphilicity caused by shifts in backbone polarity. However, the magnitude and direction of this influence depend on side-chain polarity and peptide length, making the resulting bias highly sequence-specific and difficult to anticipate or correct systematically. These findings emphasize the need to reevaluate the conventional use of extended β-sheet encoding (E-flag) and advocate for more native-like backbone representations in peptide self-assembly simulations.