Robust deep learning based protein sequence design using ProteinMPNN

By J. Dauparas, Ivan V. Anishchenko, N. Bennett, H. Bai, R. Ragotte, L. Milles, B. Wicky, A. Courbet, R. D. de Haas, N. Bethel and 12 more. bioRxiv, 2022

1,912 citations recorded by the source corpus.

While deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here we describe a deep learning based protein sequence design method, ProteinMPNN, with outstanding performance in both in silico and experimental tests. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4%, compared to 32.9% for Rosetta. Incorporation of noise during training improves sequence recovery on protein structure models, and produces sequences which more robustly encode their structures as assessed using structure prediction algorithms. We demonstrate the broad utility and high accuracy of ProteinMPNN using X-ray crystallography, cryoEM and functional studies by rescuing previously failed designs, made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target binding proteins. One-sentence summary A deep learning based protein sequence design method is described that is widely applicable to current design challenges and shows outstanding performance in both in silico and experimental tests.

Subfields: Molecular Biology, Computational Biology & Bioinformatics

View Paper: https://doi.org/10.1126/science.add2187

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

In Computational Biology & Bioinformatics

Citations in the corpus

Generalized Biomolecular Modeling and Design with RoseTTAFold All-Atom (2023, positive): “After sequencedesignusingLigandMPNN( 49 , 50 ), Rosetta GALigandDock ( 32 ) energy calculations were used to evaluate the protein – small molecule interface and AF2 predictions to evaluate the extent to which the sequence encodes the designedstructure( 45 , 51 ).”
State-specific protein–ligand complex structure prediction with a multiscale deep generative model (2022, positive): “Several works have demonstrated that deep generative models are capable of producing de novo-designed proteins with experimentally validated functions, including language models for protein sequence design 9,27,28 and diffusion models for protein backbone generation 29–32 .”
Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problem (2022, positive): “For each generated scaffold we use a C-α only version of ProteinMPNN [8] with a temperature of 0.”
Illuminating protein space with a programmable generative model (2022, background): “34).”
SaProt: Protein Language Modeling with Structure-aware Vocabulary (2024, background): “Figure 5 depicts the results of fine-tuning SaProt on the DeepLoc (Almagro Armenteros et al., 2017) dataset.”
De novo design of protein structure and function with RFdiffusion (2023, positive): “We explored two different strategies for training RFdiffusion: (1) in a manner akin to ‘canonical’ diffusion models, with predictions at each timestep independent of predictions at previous timesteps (as in previous work(5,8,9,16)), and (2) with self-conditioning(25), in which the model can conditio”
Protein structure generation via folding diffusion (2022, methodology): “We performed self-consistency TM score evaluation for both these methods, generating 8 candidate sequences using author-recommended temperature values ( T =1 : 0 for ESM-IF1, T =0 : 1 for ProteinMPNN).”
ProGen2: Exploring the Boundaries of Protein Language Models (2022, methodology): “Extensive experimental validation demonstrated that sequences generated by ProteinMPNN dramatically improved soluble expression and folded into their intended structures (Wicky et al., 2022).”
De novo protein design by inversion of the AlphaFold structure prediction network (2022, methodology): “Deep learning methods can also be used for sequence generation given a target backbone; through the interpolation between sequence and structural data, sequences that fit a target topology can be found (Anand et al., 2022; Dauparas et al., 2022; Ingraham et al., 2019).”
Masked Inverse Folding with Sequence Transfer for Protein Representation Learning (2023, background): “…parametrized as a graph neural network (GNN) (Ingra-ham et al., 2019; Strokach et al., 2020; Jin et al., 2021; Jing et al., 2020; Hsu et al., 2022; Dauparas et al., 2022; Shi et al., 2022) or SE(3)-equivariant transformer (McPartlon et al., 2022) that either reconstructs or autoregressively…”