Improving <i>de novo</i> Protein Binder Design with Deep Learning

Bennett, Nathaniel R.; Coventry, Brian; Goreshnik, Inna; Huang, Buwei; Allen, Aza; Vafeados, Dionne; Peng, Ying Po; Dauparas, Justas; Baek, Minkyung; Stewart, Lance; DiMaio, Frank; Munck, Steven De; Savvides, Savvas N.; Baker, David

doi:10.1101/2022.06.15.495993

Cited by 49 publications

(71 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ProteinMPNN can also generate binder sequences with higher success rate than previously, although it relies on specifying the backbone trace of a binder structure [14]. Reevaluating previous designs [3] using AF as a scoring function reports experimental success rates of close to 90% in some cases compared to only 0-5% using physics based calculations [14]. These methods have been applied to design binders both with and without scaffolds and known binding motifs [15].…”

Section: Introductionmentioning

confidence: 99%

“…A recent protein design method, ProteinMPNN [13], improved further on these achievements and created proteins with significantly higher solubility compared to using AF alone. ProteinMPNN can also generate binder sequences with higher success rate than previously, although it relies on specifying the backbone trace of a binder structure [14]. Reevaluating previous designs [3] using AF as a scoring function reports experimental success rates of close to 90% in some cases compared to only 0-5% using physics based calculations [14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EvoBind: in silico directed evolution of peptide binders with AlphaFold

Bryant

Elofsson

2022

Preprint

View full text Add to dashboard Cite

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

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

EvoBind: in silico directed evolution of peptide binders with AlphaFold

Bryant

Elofsson

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, the pLDDT confidence score of AlphaFold2[Jumper et al, 2021] has had a very significant impact in many applications[Necci et al, 2021;Bennett et al, 2022].4 RDKit ETKDG is a popular method for predicting the seed conformation. Although the structures may not be predicted perfectly, the errors lie largely in the torsion angles, which are resampled anyways.…”

mentioning

confidence: 99%

DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking

Corso¹,

Stark²,

Jing³

et al. 2022

Preprint

114

146

View full text Add to dashboard Cite

Predicting the binding structure of a small molecule ligand to a protein-a task known as molecular docking-is critical to drug design. Recent deep learning methods that treat docking as a regression problem have decreased runtime compared to traditional search-based methods but have yet to offer substantial improvements in accuracy. We instead frame molecular docking as a generative modeling problem and develop DIFFDOCK, a diffusion generative model over the non-Euclidean manifold of ligand poses. To do so, we map this manifold to the product space of the degrees of freedom (translational, rotational, and torsional) involved in docking and develop an efficient diffusion process on this space. Empirically, DIFFDOCK obtains a 38% top-1 success rate (RMSD<2 Å) on PDB-Bind, significantly outperforming the previous state-of-the-art of traditional docking (23%) and deep learning (20%) methods. Moreover, DIFFDOCK has fast inference times and provides confidence estimates with high selective accuracy.

show abstract

“…from the most target-compatible input fold), and the success rates at each noise scale are separated. In line with current best practice 26 , we tested using Rosetta FastRelax 52 before designing the sequence with ProteinMPNN, but found that this did not systematically improve designs. Success is defined in line with current best practice 26 : AF2 pLDDT of the monomer > 80, AF2 interaction pAE < 10, AF2 RMSD monomer vs design < 1Å.…”

Section: Supplementary Figuresmentioning

confidence: 79%

Broadly applicable and accurate protein design by integrating structure prediction networks and diffusion generative models

Watson

Juergens

Bennett

et al. 2022

Preprint

Self Cite

111

106

View full text Add to dashboard Cite

There has been considerable recent progress in designing new proteins using deep learning methods. Despite this progress, a general deep learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher order symmetric architectures, has yet to be described. Diffusion models have had considerable success in image and language generative modeling but limited success when applied to protein modeling, likely due to the complexity of protein backbone geometry and sequence-structure relationships. Here we show that by fine tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding, and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold Diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of new designs. In a manner analogous to networks which produce images from user-specified inputs, RFdiffusion enables the design of diverse, complex, functional proteins from simple molecular specifications.

show abstract

Improving de novo Protein Binder Design with Deep Learning

Cited by 49 publications

References 16 publications

EvoBind: in silico directed evolution of peptide binders with AlphaFold

EvoBind: in silico directed evolution of peptide binders with AlphaFold

DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking

Broadly applicable and accurate protein design by integrating structure prediction networks and diffusion generative models

Contact Info

Product

Resources

About