Title: Learning structural motif representations for efficient protein structure search
AbstractMotivation
Given a protein of unknown function, fast identification of similar protein structures from the Protein Data Bank (PDB) is a critical step for inferring its biological function. Such structural neighbors can provide evolutionary insights into protein conformation, interfaces and binding sites that are not detectable from sequence similarity. However, the computational cost of performing pairwise structural alignment against all structures in PDB is prohibitively expensive. Alignment-free approaches have been introduced to enable fast but coarse comparisons by representing each protein as a vector of structure features or fingerprints and only computing similarity between vectors. As a notable example, FragBag represents each protein by a ‘bag of fragments’, which is a vector of frequencies of contiguous short backbone fragments from a predetermined library. Despite being efficient, the accuracy of FragBag is unsatisfactory because its backbone fragment library may not be optimally constructed and long-range interacting patterns are omitted.
Results
Here we present a new approach to learning effective structural motif presentations using deep learning. We develop DeepFold, a deep convolutional neural network model to extract structural motif features of a protein structure. We demonstrate that DeepFold substantially outperforms FragBag on protein structural search on a non-redundant protein structure database and a set of newly released structures. Remarkably, DeepFold not only extracts meaningful backbone segments but also finds important long-range interacting motifs for structural comparison. We expect that DeepFold will provide new insights into the evolution and hierarchical organization of protein structural motifs.
We present an approach for the efficient docking of peptide motifs to their free receptor structures. Using a motif based search, we can retrieve structural fragments from the Protein Data Bank (PDB) that are very similar to the peptide’s final, bound conformation. We use a Fast Fourier Transform (FFT) based docking method to quickly perform global rigid body docking of these fragments to the receptor. According to CAPRI peptide docking criteria, an acceptable conformation can often be found among the top-ranking predictions.
Availability and Implementation
The method is available as part of the protein-protein docking server ClusPro at https://peptidock.cluspro.org/nousername.php.
Supplementary information
Supplementary data are available at Bioinformatics online.
Cao, Yiwei; Park, Sang-Jun; Mehta, Akul Y.; Cummings, Richard D.; Im, Wonpil; Elofsson, ed., Arne(
, Bioinformatics)
AbstractMotivation
Glycan microarrays are capable of illuminating the interactions of glycan-binding proteins (GBPs) against hundreds of defined glycan structures, and have revolutionized the investigations of protein–carbohydrate interactions underlying numerous critical biological activities. However, it is difficult to interpret microarray data and identify structural determinants promoting glycan binding to glycan-binding proteins due to the ambiguity in microarray fluorescence intensity and complexity in branched glycan structures. To facilitate analysis of glycan microarray data alongside protein structure, we have built the Glycan Microarray Database (GlyMDB), a web-based resource including a searchable database of glycan microarray samples and a toolset for data/structure analysis.
Results
The current GlyMDB provides data visualization and glycan-binding motif discovery for 5203 glycan microarray samples collected from the Consortium for Functional Glycomics. The unique feature of GlyMDB is to link microarray data to PDB structures. The GlyMDB provides different options for database query, and allows users to upload their microarray data for analysis. After search or upload is complete, users can choose the criterion for binder versus non-binder classification. They can view the signal intensity graph including the binder/non-binder threshold followed by a list of glycan-binding motifs. One can also compare the fluorescence intensity data from two different microarray samples. A protein sequence-based search is performed using BLAST to match microarray data with all available PDB structures containing glycans. The glycan ligand information is displayed, and links are provided for structural visualization and redirection to other modules in GlycanStructure.ORG for further investigation of glycan-binding sites and glycan structures.
Availability and implementation
http://www.glycanstructure.org/glymdb.
Contact
wonpil@lehigh.edu
Supplementary information
Supplementary data are available at Bioinformatics online.
Zhang, Chengxin; Zheng, Wei; Mortuza, S M; Li, Yang; Zhang, Yang; Valencia, Alfonso(
, Bioinformatics)
Abstract Motivation The success of genome sequencing techniques has resulted in rapid explosion of protein sequences. Collections of multiple homologous sequences can provide critical information to the modeling of structure and function of unknown proteins. There are however no standard and efficient pipeline available for sensitive multiple sequence alignment (MSA) collection. This is particularly challenging when large whole-genome and metagenome databases are involved. Results We developed DeepMSA, a new open-source method for sensitive MSA construction, which has homologous sequences and alignments created from multi-sources of whole-genome and metagenome databases through complementary hidden Markov model algorithms. The practical usefulness of the pipeline was examined in three large-scale benchmark experiments based on 614 non-redundant proteins. First, DeepMSA was utilized to generate MSAs for residue-level contact prediction by six coevolution and deep learning-based programs, which resulted in an accuracy increase in long-range contacts by up to 24.4% compared to the default programs. Next, multiple threading programs are performed for homologous structure identification, where the average TM-score of the template alignments has over 7.5% increases with the use of the new DeepMSA profiles. Finally, DeepMSA was used for secondary structure prediction and resulted in statistically significant improvements in the Q3 accuracy. It is noted that all these improvements were achieved without re-training the parameters and neural-network models, demonstrating the robustness and general usefulness of the DeepMSA in protein structural bioinformatics applications, especially for targets without homologous templates in the PDB library. Availability and implementation https://zhanglab.ccmb.med.umich.edu/DeepMSA/. Supplementary information Supplementary data are available at Bioinformatics online.
Di Costanzo, Luigi; Dutta, Shuchismita; Burley, Stephen K.(
, Biopolymers)
Abstract
Extensive efforts invested in understanding the rules of protein folding are now being applied, with good effect, in de novo design of proteins/peptides. For proteins containing standard α‐amino acids alone, knowledge derived from experimentally determined three‐dimensional (3D) structures of proteins and biologically active peptides are available from the Protein Data Bank (PDB), and the Cambridge Structural Database (CSD). These help predict and design protein structures, with reasonable confidence. However, our knowledge of 3D structures of biomolecules containing backbone modified amino acids is still evolving. A major challenge in de novo protein/peptide design concerns the engineering of conformationally constrained molecules with specific structural elements and chemical groups appropriately positioned for biological activity. This review explores four classes of amino acid modifications that constrain protein/peptide backbone structure. Systematic analysis of peptidic molecule structures (eg, bioactive peptides, inhibitors, antibiotics, and designed molecules), containing these backbone‐modified amino acids, found in the PDB and CSD are discussed. The review aims to provide structure–function insights that will guide future design of proteins/peptides.
The state-of-art protein structure prediction methods such as AlphaFold are being widely used to predict structures of uncharacterized proteins in biomedical research. There is a significant need to further improve the quality and nativeness of the predicted structures to enhance their usability. In this work, we develop ATOMRefine, a deep learning-based, end-to-end, all-atom protein structural model refinement method. It uses a SE(3)-equivariant graph transformer network to directly refine protein atomic coordinates in a predicted tertiary structure represented as a molecular graph.
Results
The method is first trained and tested on the structural models in AlphaFoldDB whose experimental structures are known, and then blindly tested on 69 CASP14 regular targets and 7 CASP14 refinement targets. ATOMRefine improves the quality of both backbone atoms and all-atom conformation of the initial structural models generated by AlphaFold. It also performs better than two state-of-the-art refinement methods in multiple evaluation metrics including an all-atom model quality score—the MolProbity score based on the analysis of all-atom contacts, bond length, atom clashes, torsion angles, and side-chain rotamers. As ATOMRefine can refine a protein structure quickly, it provides a viable, fast solution for improving protein geometry and fixing structural errors of predicted structures through direct coordinate refinement.
Availability and implementation
The source code of ATOMRefine is available in the GitHub repository (https://github.com/BioinfoMachineLearning/ATOMRefine). All the required data for training and testing are available at https://doi.org/10.5281/zenodo.6944368.
@article{osti_10421939,
place = {Country unknown/Code not available},
title = {Learning structural motif representations for efficient protein structure search},
url = {https://par.nsf.gov/biblio/10421939},
DOI = {10.1093/bioinformatics/bty585},
abstractNote = {Abstract MotivationGiven a protein of unknown function, fast identification of similar protein structures from the Protein Data Bank (PDB) is a critical step for inferring its biological function. Such structural neighbors can provide evolutionary insights into protein conformation, interfaces and binding sites that are not detectable from sequence similarity. However, the computational cost of performing pairwise structural alignment against all structures in PDB is prohibitively expensive. Alignment-free approaches have been introduced to enable fast but coarse comparisons by representing each protein as a vector of structure features or fingerprints and only computing similarity between vectors. As a notable example, FragBag represents each protein by a ‘bag of fragments’, which is a vector of frequencies of contiguous short backbone fragments from a predetermined library. Despite being efficient, the accuracy of FragBag is unsatisfactory because its backbone fragment library may not be optimally constructed and long-range interacting patterns are omitted. ResultsHere we present a new approach to learning effective structural motif presentations using deep learning. We develop DeepFold, a deep convolutional neural network model to extract structural motif features of a protein structure. We demonstrate that DeepFold substantially outperforms FragBag on protein structural search on a non-redundant protein structure database and a set of newly released structures. Remarkably, DeepFold not only extracts meaningful backbone segments but also finds important long-range interacting motifs for structural comparison. We expect that DeepFold will provide new insights into the evolution and hierarchical organization of protein structural motifs. Availability and implementationhttps://github.com/largelymfs/DeepFold},
journal = {Bioinformatics},
volume = {34},
number = {17},
publisher = {Oxford University Press},
author = {Liu, Yang and Ye, Qing and Wang, Liwei and Peng, Jian},
}
Warning: Leaving National Science Foundation Website
You are now leaving the National Science Foundation website to go to a non-government website.
Website:
NSF takes no responsibility for and exercises no control over the views expressed or the accuracy of
the information contained on this site. Also be aware that NSF's privacy policy does not apply to this site.
