PepNN: a deep attention model for the identification of peptide binding sites

Abdin, Osama; Wen, Han; Kim, Philip M.

doi:10.1101/2021.01.10.426132

Cited by 8 publications

(12 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We additionally compared CAMP with other methods on several representative benchmark data sets (Supplementary Table 4) that were originally used to evaluate the performance of peptide docking and detecting "hotspots" at protein interface 34,[39][40][41][42] . As shown in Supplementary Fig.…”

Section: New Insights By Characterizing Binding Residues On Peptidesmentioning

confidence: 99%

A deep-learning framework for multi-level peptide–protein interaction prediction

et al. 2021

View full text Add to dashboard Cite

Peptide-protein interactions are involved in various fundamental cellular functions and their identification is crucial for designing efficacious peptide therapeutics. Recently, a number of computational methods have been developed to predict peptide-protein interactions. However, most of the existing prediction approaches heavily depend on high-resolution structure data. Here, we present a deep learning framework for multi-level peptide-protein interaction prediction, called CAMP, including binary peptide-protein interaction prediction and corresponding peptide binding residue identification. Comprehensive evaluation demonstrated that CAMP can successfully capture the binary interactions between peptides and proteins and identify the binding residues along the peptides involved in the interactions. In addition, CAMP outperformed other state-of-the-art methods on binary peptide-protein interaction prediction. CAMP can serve as a useful tool in peptide-protein interaction prediction and identification of important binding residues in the peptides, which can thus facilitate the peptide drug discovery process.

show abstract

Section: New Insights By Characterizing Binding Residues On Peptidesmentioning

confidence: 99%

A deep-learning framework for multi-level peptide–protein interaction prediction

et al. 2021

View full text Add to dashboard Cite

show abstract

“…We did not observe inferior performance metrics, indicating that there is no bias (see Figure S2). Unfortunately, we could not calculate these types of results for the other methods, but for PepNN, that shows a similar trend (see Figure S2).…”

Section: Resultsmentioning

confidence: 96%

“…or structural (e.g., solvent-accessible surface area, secondary structure type, etc.) descriptors fed into a machine learning model to score each amino acid for belonging to a binding site. − End-to-end methods operate directly with protein sequences , or structures by taking advantage of deep learning methods capable of learning features during training. With a continuously growing amount of structural data, it becomes possible to develop more robust methods using end-to-end deep learning approaches that work directly with spatial structures of protein complexes.…”

Section: Introductionmentioning

confidence: 99%

Protein–Peptide Binding Site Detection Using 3D Convolutional Neural Networks

Kozlovskii

Popov

2021

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Peptides and peptide-based molecules represent a promising therapeutic modality targeting intracellular protein−protein interactions, potentially combining the beneficial properties of biologics and small-molecule drugs. Protein−peptide complexes occupy a unique niche of interaction interfaces with respect to protein−protein and protein− small molecule complexes. Protein−peptide binding site identification resembles image object detection, a field that had been revolutionalized with computer vision techniques. We present a new protein−peptide binding site detection method called BiteNet Pp by harnessing the power of 3D convolutional neural network. Our method employs a tensor-based representation of spatial protein structures, which is fed to 3D convolutional neural network, resulting in probability scores and coordinates of the binding "hot spots" in the input structures. We used the domain adaptation technique to fine-tune model trained on protein−small molecule complexes using a manually curated set of protein−peptide structures. BiteNet Pp consistently outperforms existing state-of-the-art methods in the independent test benchmark. It takes less than a second to analyze a single-protein structure, making BiteNet Pp suitable for the large-scale analysis of protein− peptide binding sites.

show abstract

“…The model architecture untilized in GDockScore builds upon the parallel embedding architecture utilized by Abdin et al (Abdin et al 2021) and protein graph attention developed by Ingraham et al (Ingraham et al n.d.). Details of the model input representation are available in the Methods section.…”

Section: Resultsmentioning

confidence: 99%

GDockScore: a graph-based protein-protein docking scoring function

McFee

Kim

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Protein complexes play vital roles in a variety of biological processes such as mediating biochemical reactions, the immune response, and cell signalling, with three-dimensional structure specifying function. Computational docking methods provide a means to determine the interface between two complexed polypeptide chains without using time-consuming experimental techniques. The docking process requires the optimal solution to be selected with a scoring function. Here we propose a novel graph-based deep learning model that utilizes mathematical graph representations of proteins to learn a scoring function (GDockScore). GDockScore was pre-trained on docking outputs generated with the Protein Data Bank (PDB) biounits and the RosettaDock protocol, and then fine-tuned on HADDOCK decoys generated on the ZDOCK Protein Docking Benchmark. GDockScore performs similarly to the Rosetta scoring function on docking decoys generated using the RosettaDock protocol. Furthermore, state-of-the-art is achieved on the CAPRI score set, a challenging dataset for developing docking scoring functions.

show abstract

PepNN: a deep attention model for the identification of peptide binding sites

Cited by 8 publications

References 59 publications

A deep-learning framework for multi-level peptide–protein interaction prediction

A deep-learning framework for multi-level peptide–protein interaction prediction

Protein–Peptide Binding Site Detection Using 3D Convolutional Neural Networks

GDockScore: a graph-based protein-protein docking scoring function

Contact Info

Product

Resources

About