Mirror score improves ranking in designing protease resistant d peptides
Mirror-Score: Calibrated, Inference-only Scoring Exposes the Limits of Sequence-compatibility Ranking in D-peptide Design
Artificial IntelligenceComputational Engineering, Finance, and ScienceEmerging TechnologiesMachine Learning
Summary
Designing special peptides made of D-amino acids can help create drugs that stay longer in the body and stick better to their targets. The authors show that a previous method to rank these designed peptides by how well they bind doesn’t work reliably. They introduce Mirror-Score, a new way to score and rank these peptides that better matches real binding data. Their method works well within specific protein families but needs more data to work broadly across different types.
What this means in practice
- •For protein engineers: Use Mirror-Score to better predict which designed D-peptides will bind strongly to target proteins within specific families.
- •For drug discovery teams: Improve computational screening of stable D-peptide drug candidates by applying family-specific calibration of ranking scores.
Authors
Jiada Li
Abstract
D-peptides combine protease resistance with high target specificity, but computational design of D-peptide binders remains immature. Mirror-Peptidizer introduced an in silico mirror-image screening pipeline using target reflection, backbone generation, and ProteinMPNN sequence design, but its raw ProteinMPNN negative log-likelihood (NLL) ranking was not validated against measured affinities, and only 4 of 9 tested MDM2 designs bound detectably. We introduce Mirror-Score, a calibrated, inference-only scoring framework for heterochiral D-peptide/L-protein complexes, and a public benchmark of 31 crystal complexes across four target families, including 18 with literature-verified affinities. Raw ProteinMPNN NLL is not a valid affinity ranker: its pooled Spearman correlation with affinity is 0.19, and correlations reverse between MDM2/CHIP (+0.62) and gp41 (-0.70). We therefore evaluate Boltz-2 mirror-space cofolding confidence. For the complete viral-entry family (7 structures representing 3 peptides), interface predicted local distance difference test (pLDDT) achieves structure-level leave-one-out Spearman rho = 0.90 (p = 0.006) and correctly orders all three peptides by affinity, whereas NLL fails (structure-level rho = 0.18). Because only three independent chemotypes are represented, this result indicates directional consistency rather than a statistically validated predictor. Cross-family calibration does not transfer at current sample sizes, supporting family-matched calibration as the practical deployment mode. We also specify a prospective design protocol for the antimicrobial-resistance targets LasR and LecB from Pseudomonas aeruginosa, including mirrored structures, ligand-derived hotspot maps, diffusion-model-ready inputs, and Mirror-Score ranking. Code, benchmark data, structures, and analysis scripts are openly available at https://github.com/Jiadalee/Mirror-Score.