Structure-Guided Peptide Discovery and Optimization

Receptor.AI will present at the PDHC De Novo Design Webinar on September 29, 2026, at 11:00 AM ET. CEO Alan Nafiiev and CSO Semen Yesylevskyy will introduce the Receptor.AI approach to structure-guided peptide discovery and optimization.

Peptide discovery projects often begin with incomplete information. Experimental screening technologies such as mRNA or phage display can identify active peptide sequences without revealing where or how they bind to the target. For other proteins, these screening approaches may be difficult to implement at all, leaving the project without a practical starting peptide. Once a binder is available, optimisation introduces a further challenge: sequence changes must improve molecular recognition while maintaining the physicochemical, developability, and project-specific properties required of a therapeutic lead.

In this webinar, we will present a structure-guided peptide design platform built to address these challenges across the discovery workflow. The platform can identify binding hotspots on protein surfaces, determine plausible binding modes for experimentally discovered peptides, and construct peptide–protein complexes for subsequent design. When suitable experimental hits are not available, de novo peptide generation can be performed directly against structurally defined regions of the target. Candidates originating from mRNA display, phage display, external generative approaches, or other discovery routes can therefore be brought into the same structure-based workflow.

Once a peptide–protein complex has been established, optimization is performed in the context of the target rather than at the sequence level alone. AI-guided design is combined with physics-based molecular modelling and scoring to evaluate how proposed modifications affect binding geometry and interaction quality within the complex. This enables multiparameter optimization across binding, physicochemical properties, developability objectives, chemical and sequence constraints, and other project-specific requirements, while focusing experimental testing on the most promising designs.

The session will conclude with a demonstration of the platform, showing how an initial peptide hit can progress through binding-site and binding-mode identification, structural evaluation, and iterative optimization toward higher-quality peptide leads.