Protein sequence design with a learned potential

By Namrata Anand-Achim, Raphael R. Eguchi, Alexander Derry, Russ B. Altman, Po-Ssu Huang. 2020

135 citations recorded by the source corpus.

The primary challenge of fixed-backbone protein design is to find a distribution of sequences that fold to the backbone of interest. This task is central to nearly all protein engineering problems, as achieving a particular backbone conformation is often a prerequisite for hosting specific functions. In this study, we investigate the capability of a deep neural network to learn the requisite patterns needed to design sequences. The trained model serves as a potential function defined over the space of amino acid identities and rotamer states, conditioned on the local chemical environment at each residue. While most deep learning based methods for sequence design only produce amino acid sequences, our method generates full-atom structural models, which can be evaluated using established sequence quality metrics. Under these metrics we are able to produce realistic and variable designs with quality comparable to the state-of-the-art. Additionally, we experimentally test designs for a de novo TIM-barrel structure and find designs that fold, demonstrating the algorithm’s generalizability to novel structures. Overall, our results demonstrate that a deep learning model can match state-of-the-art energy functions for guiding protein design. Significance Protein design tasks typically depend on carefully modeled and parameterized heuristic energy functions. In this study, we propose a novel machine learning method for fixed-backbone protein sequence design, using a learned neural network potential to not only design the sequence of amino acids but also select their side-chain configurations, or rotamers. Factoring through a structural representation of the protein, the network generates designs on par with the state-of-the-art, despite having been entirely learned from data. These results indicate an exciting future for protein design driven by machine learning.

Subfields: Materials Chemistry, Molecular Biology, Computational Biology & Bioinformatics

How researchers see this paper

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

In Materials Chemistry

In Molecular Biology

In Computational Biology & Bioinformatics

Citations in the corpus

Protein Structure and Sequence Generation with Equivariant Denoising Diffusion Probabilistic Models (2022, methodology): “The 3DConv baseline refers to a machine learning approach for sequence design and rotamer packing using 3D convolutions [3].”
Learning from Protein Structure with Geometric Vector Perceptrons (2020, methodology): “A number of CPD methods (Anand et al., 2020; Zhang et al., 2019; Shroff et al., 2019) and the MQA methods 3DCNN (Derevyanko et al., 2018) and Ornate (Pag ` es et al., 2019) exemplify the power of this approach.”
Protein sequence design by conformational landscape optimization (2021, background): “…landscape was quantified with P near (28), which approximates the Boltzmann-weighted probability of the structure adapting the target conformation (fuzzy cutoff): , [6] where N is the number of decoys, E is the energy of the decoy, and RMSD represents its structural deviation from the target state.”
Deep Learning in Protein Structural Modeling and Design (2020, methodology): “Anand et al.202 trained a similar model to design sequences for a given backbone.”
ProtGPT2 is a deep unsupervised language model for protein design (2022, background): “In particular, 1068 shows a TIM-barrel fold, a topology which to date has met impressive success in de novo design 62 – 64 , but whose idealized structure has nevertheless proven challenging to extend via additional secondary elements and longer loops 65,66 .”
Protein sequence design with deep generative models (2021, background): “Additionally, while studies incorporating knowledge of protein structure are becoming increasingly powerful [9, 10, 11, 12, 13], they are beyond the scope of this review, and we focus on deep generative models of protein sequence.”
ATOM3D: Tasks On Molecules in Three Dimensions (2020, methodology): “, 2018], protein sequence design [Anand et al., 2020], protein interface prediction [Townshend et al.”
AlphaDesign: A de novo protein design framework based on AlphaFold (2021, methodology): “[ 64, 65, 67] have used per-residue local coordinate systems to reason about protein geometry, while [66] use network architecture adapted to Euclidean coordinate data.”
Deep learning techniques have significantly impacted protein structure prediction and protein design. (2021, methodology): “Using a different approach, Anand et al. recently extracted features around a local, voxelized environment for each residue, which were then fed into a convolutional neural network with six 3D convolutional layers [63].”