EvoBind: in silico directed evolution of peptide binders with AlphaFold

Abstract

Currently, there is no accurate method to computationally design peptide binders towards a specific protein interface using only a target structure. Experimental methods such as phage display can produce strong binders, but it is impossible to know where these bind without solving the structures. Using AlphaFold2 (AF) and other AI methods to distinguish true binders has proven highly successful but relies on the availability of binding scaffolds. Here, we develop EvoBind, an in silico directed-evolution platform based on AF that designs peptide binders towards an interface using only sequence information. We show that AF can distinguish between native and mutated peptide binders using the plDDT score and find that AF adapts the receptor interface structure to the binders during optimisation. We analyse previously designed minibinder proteins and show that AF can distinguish designed binders from non-binders. We compare ELISA ratios of different peptide binders and find the affinity can not be distinguished among binders, possibly due to varying binding sites and low AF confidence. We test the recovery of binding motifs and find that up to 75% of motifs are recovered. In principle, EvoBind can be used to design binders towards any interface conditioned on if AF can predict these. We expect that EvoBind will aid experimentalists substantially, providing a starting point for further laboratory analysis and optimisation. We hope that the use of AI-based methods will come to make binder design significantly cheaper and more accurate in tackling unmet clinical needs. EvoBind is freely available at: https://colab.research.google.com/github/patrickbryant1/EvoBind/blob/master/EvoBind.ipynb

Funneling modulatory peptide design with generative models: Discovery and characterization of disruptors of calcineurin protein-protein interactions

“…Accordingly, we found that our cRBM model better predicted binding affinity changes upon mutation that ProteinMPNN [58], a recently published structure-based autoregressive graph neural network for sequence design. Synergy between evolutionary-based and structure-based protein design approaches is well-established [68][69][70][71] and therefore, although structure-based computational design methods are rapidly improving [58,72,73], we expect that evolutionary information will still prove valuable in the future.…”

Multiplexed selectivity screening of anti-GPCR antibodies

“…The scope of machine learning applications has expanded dramatically within the past three years and accelerated the prediction of protein structures. The advent of Alpha-Fold ( 22 ) has recently been harnessed by others to create in silico models for designing binding reagents for any interface ( 24 ). It will be interesting to follow the development of such approaches and compare the experimental validation data accompanying computational classification schemes.…”

“…The advent of Alpha-Fold (22) has recently been harnessed by others to create in silico models for designing binding reagents for any interface (24). It will be interesting to follow the development of such approaches and compare the experimental validation data accompanying computational classification schemes.…”

Identifying endogenous peptide receptors by combining structure and transmembrane topology prediction

“…Note that we do not necessarily expect the receptor to have the top rank -many peptides are known to have more than one endogenous binding partner [6]. Moreover, the exact ordering of receptors by AlphaFold confidence is unlikely to correlate to binding strength [19]. However, with respect to a full receptor library, any true receptor can still be expected to rank high.…”