An end-to-end deep learning method for protein side-chain packing and inverse folding

McPartlon, Matt; Xu, Jinbo

doi:10.1073/pnas.2216438120

Cited by 21 publications

(26 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To further assess the balance between high-quality data and dataset size for PSCP, we additionally experimented with training models on the BC40 data 35 , which has been used recently for training data in PSCP methods 24,27 . This dataset consists of 36,970 protein chains released before August 2020 that are nonredundant at 40% sequence identity but have no other filters.…”

Section: Bc40 Dataset Preparationmentioning

confidence: 99%

“…Graph neural networks (GNNs) have shown remarkable promise in modeling proteins and have been successfully applied to various protein tasks, including fold classification 40,41 , property prediction [40][41][42] , fixed backbone sequence design 27,[43][44][45][46] , and PSCP 24,[27][28][29] . With the rationale that the specific side chain conformations are primarily dependent upon the local environment of the amino acid, we decided to model the PSCP problem with a GNN, wherein each residue is modeled as a node and is connected to its 𝑘 nearest neighbors.…”

Section: Architectural Considerationsmentioning

confidence: 99%

“…While these methods produce coordinates for all heavy atoms in the protein and are, therefore, capable of side chain packing, there is no way to pre-specify and hold the backbone conformation fixed to mimic the PSCP task. On the other hand, DL-based methods designed specifically for PSCP have shown significant accuracy improvements over traditional approaches, while often being less time-consuming [23][24][25][26][27][28][29] . These methods either directly predict the location of side chain atoms 25,27 or torsion angles from which the side chain can be reconstructed with idealized geometry 23,24,28 .…”

Section: Introductionmentioning

confidence: 99%

“…Building off the recent success of graph neural networks for encoding and propagating structural information within proteins 27,[30][31][32][33] , we present the Protein Invariant Point Packer (PIPPack) which can rapidly and accurately predict side chain conformations. After investigating the balance between data quality and dataset size, we trained our final model on non-redundant subset of the Protein Data Bank (PDB) 30 to jointly predict binned 𝜒 dihedral angles for each residue, refining its previous predictions with recycling and explicitly incorporating, throughout the network, the geometry of the protein backbone through a novel message passing scheme.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Invariant point message passing for protein side chain packing

Randolph

Kuhlman

2023

Preprint

View full text Add to dashboard Cite

Protein side chain packing (PSCP) is a fundamental problem in the field of protein engineering, as high confidence and low energy conformations of amino acid side chains are crucial for understanding (and designing) protein folding, protein-protein interactions, and protein-ligand interactions. Traditional PSCP methods (such as the Rosetta Packer) often rely on a library of discrete side chain conformations, or rotamers, and a forcefield to guide the structure to low energy conformations. Recently, deep learning (DL) based methods (such as DLPacker, AttnPacker, and DiffPack) have demonstrated state-of-the-art predictions and speed in the PSCP task. Building off the success of graph and message passing neural networks for protein modeling, we present the Protein Invariant Point Packer (PIPPack) which effectively processes local structural and sequence information to produce realistic, idealized side chain coordinates using 𝜒-angle distribution predictions and geometry-aware invariant point message passing (IPMP). To demonstrate its broad applicability to protein-related tasks, IPMP was additionally incorporated in a fixed backbone protein design method, which enabled the generation of more native-like sequences than common message passing schemes. On a test set of ∼1,400 high-quality protein chains, PIPPack performs competitively with other state-of-the-art PSCP methods in rotamer recovery and per-residue RMSDs but is significantly faster.

show abstract

Section: Bc40 Dataset Preparationmentioning

confidence: 99%

Section: Architectural Considerationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Invariant point message passing for protein side chain packing

Randolph

Kuhlman

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…More recently, machine learning approaches have significantly advanced protein structural analysis research [6][7][8][9][10][11] . Among these, the most impressive is the Google Deep-Mind's AlphaFold AI program 12 , which has won the Casp and Lasker basic medical research awards for predicting (with circa 95% accuracy) protein's 3D structures from their amino-acid sequences.…”

mentioning

confidence: 99%

Discovering Secondary Protein Structures via Local Euler Curvature

Moreira,

Braddell,

Santos

et al. 2023

Preprint

View full text Add to dashboard Cite

Protein structure analysis and classification, which is fundamental for predicting protein function, still poses formidable challenges in the fields of molecular biology, mathematics, physics and computer science. In the present work we exploit recent advances in computational topology to define a new intrinsic unsupervised topological fingerprint for proteins. These fingerprints, computed via Local Euler Curvature (LECs), identify secondary protein structures, such as Helices and Sheets, by capturing their distinctive topological signatures. Using an extensive protein residue database, the proposed computational framework not only distinguishes between structural classes via unsupervised clustering but also achieves remarkable accuracy in classifying proteins structures through supervised machine learning classifier. We also show that the internal structure of LEC space embeds the information about the secondary structure of proteins. Beyond its immediate implications for the advancement of critical application areas such as drug design and biotechnology, our approach opens a fascinating avenue towards characterizing the multiscale structures of diverse biopolymers based solely on their geometric and topological attributes.

show abstract

Invariant point message passing for protein side chain packing

Randolph,

Kuhlman

2024

Proteins

View full text Add to dashboard Cite

Protein side chain packing (PSCP) is a fundamental problem in the field of protein engineering, as high‐confidence and low‐energy conformations of amino acid side chains are crucial for understanding (and designing) protein folding, protein–protein interactions, and protein‐ligand interactions. Traditional PSCP methods (such as the Rosetta Packer) often rely on a library of discrete side chain conformations, or rotamers, and a forcefield to guide the structure to low‐energy conformations. Recently, deep learning (DL) based methods (such as DLPacker, AttnPacker, and DiffPack) have demonstrated state‐of‐the‐art predictions and speed in the PSCP task. Building off the success of geometric graph neural networks for protein modeling, we present the Protein Invariant Point Packer (PIPPack) which effectively processes local structural and sequence information to produce realistic, idealized side chain coordinates using ‐angle distribution predictions and geometry‐aware invariant point message passing (IPMP). On a test set of ∼1400 high‐quality protein chains, PIPPack is highly competitive with other state‐of‐the‐art PSCP methods in rotamer recovery and per‐residue RMSD but is significantly faster.

show abstract

An end-to-end deep learning method for protein side-chain packing and inverse folding

Cited by 21 publications

References 51 publications

Invariant point message passing for protein side chain packing

Invariant point message passing for protein side chain packing

Discovering Secondary Protein Structures via Local Euler Curvature

Invariant point message passing for protein side chain packing

Contact Info

Product

Resources

About