Fragment-based growth models exploit pre-existing structures but can become constrained by early autoregressive decisions and generally lack knowledge of the final target shape. Here, we tested whether geometric guidance applied to a pretrained DiffSBDD model could steer generation toward the shape of a ligand B while preserving a warhead from ligand A. One-shot guidance increased target-volume coverage but predominantly produced disconnected structures. Reformulating the task as smaller scaffold expansions substantially improved the recovery of connected anchor components. Target-directed selection and adaptive growth increments enabled progressive movement toward B, while beam search improved endpoint recovery. Soft-scaffold inpainting provided graded control over the mobility of the inherited scaffold while preserving the fixed warhead. Together, these results show that inference-time shape guidance can be converted from disconnected volume coverage into controlled staged molecular growth, although performance remains dependent on the geometric compatibility of the ligand pair.

Machine Learning frameworks have emerged as a promising tool for antimicrobial peptide design; however, generative models remain limited by two persistent problems: the limited availability of experimentally validated peptides and the circular dependency of the models. In this work we present a conditional variational autoencoder pipeline that addresses both limitations through a modular architecture that combines both binary and quantitative experimental data and implements a multimodal approach to externally guide the generation. A transformer-based encoder successfully generated a discriminative 64-dimensional latent space (test AUROC 0.968, F1 0.919) separating antimicrobial from non-antimicrobial sequences. This latent representation conditions a species-specific LoRA fine-tuned ProtGPT2 decoder through a scalar gating function, which generates balanced antimicrobial peptides through two different modes; prior and perturb, depending on their generation starting points. We introduced a Surrogate Weighted Fine-Tuning (SWF) ensemble to eliminate the circular dependency and a Cross-Entropy Method to explore and exploit the latent space, leading to successful antimicrobial peptide generation. The best candidates exhibited competitive physicochemical characteristics, a mean helical fraction of 0.874 (mean pLDDT 83.7), and externally predicted efficacy evaluated by APEX.

Background/Objectives: Antimicrobial peptides (AMPs) are promising candidates against multidrug-resistant bacteria, although their clinical translation is frequently limited by cytotoxicity. In this study, we investigated whether combinations of structurally related oligomeric analogs could cooperatively enhance bacterial membrane targeting while redistributing the associated cytotoxic burden. Methods: Monomeric, dimeric, and tetrameric AMPs were evaluated through antimicrobial susceptibility testing, checkerboard interaction assays, RAW 264.7 macrophage cytotoxicity assays, and all-atom molecular dynamics simulations, including biased membrane insertion and umbrella sampling analyses. In addition, we introduced the Combinatorial Therapeutic Index (CTI) as an exploratory metric to estimate the predicted reduction in cytotoxic burden associated with peptide combinations. Results: Cytotoxicity varied substantially among oligomeric forms, with larger and more hydrophobic peptides, particularly tetramers, exhibiting the highest cytotoxicity. Additive effects were observed in checkerboard assays involving linear, dimeric, and tetrameric forms, supporting the redistribution of the toxic burden and enabling the beneficial membrane-interaction properties of hydrophobic linear peptides to be leveraged at lower cytotoxic cost. Predicted therapeutic improvement ranged from approximately twofold for the SET-M33:L33 combination to nearly ninefold for the SET-M33:DIM-33:L8 triple combination. Molecular dynamics simulations revealed non-redundant membrane interaction behaviors, with smaller peptides exhibiting deeper membrane insertion and the dimeric form favoring interfacial membrane engagement. Conclusions: These findings support a cooperative formulation strategy in which structurally related SET-M33 oligomers contribute complementary antibacterial functions while reducing the predicted cytotoxic burden. Further experimental validation using direct cytotoxicity assays of complete peptide mixtures will be necessary to confirm the therapeutic potential of these formulations.