Design of linear and cyclic peptide binders of different lengths only from a protein target sequence

Structure prediction technology has revolutionised the field of protein design, but key questions such as how to design new functions remain. Many proteins exert their functions through interactions with other proteins, and a significant challenge is designing these interactions effectively. While most efforts have focused on larger, more stable proteins, shorter peptides offer advantages such as lower manufacturing costs, reduced steric hindrance, and the ability to traverse cell membranes when cyclized.

Here, we present an AI method to design novel linear and cyclic peptide binders of varying lengths based solely on a protein target sequence. Our approach does not specify a binding site or the length of the binder, making the procedure completely blind. We demonstrate that high-affinity binders can be selected directly from predicted confidence metrics, and adversarial designs can be avoided through orthogonalin silicoevaluation, tripling the success rate.

We selected a single designed linear sequence for lengths ranging from 8 to 20 residues and evaluated the affinity using surface plasmon resonance (SPR). Of the sequences tested, 6 out of 13 (46%) displayed affinity with dissociation constants (Kd) in the micromolar range; the strongest binder had a Kd of 19 nM, and the weakest had a Kd of 7.9 μM. Our protocol,EvoBind2(<ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="https://github.com/patrickbryant1/EvoBind">https://github.com/patrickbryant1/EvoBind</ext-link>), enables binder design based solely on a protein target sequence, suggesting the potential for a rapid increase in the number of proteins that can be targeted for various biotechnological applications.