Prediction of Protein–Protein Interaction Sites Using Convolutional Neural Network and Improved Data Sets

Xie, Zengyan; Deng, Xiaoya; Shu, Kunxian

doi:10.3390/ijms21020467

Cited by 49 publications

(34 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a subfield of machine learning approaches, deep learning methods have been shown to exhibit unprecedented performance in various areas of biological prediction [51][52][53][54][55][56][57][58][59][60][61] . We described a novel deep neural network model in the present study, termed AptaNet, for predicting API.…”

Section: Discussionmentioning

confidence: 99%

AptaNet as a deep learning approach for aptamer–protein interaction prediction

Emami

Ferdousi

2021

Sci Rep

View full text Add to dashboard Cite

Aptamers are short oligonucleotides (DNA/RNA) or peptide molecules that can selectively bind to their specific targets with high specificity and affinity. As a powerful new class of amino acid ligands, aptamers have high potentials in biosensing, therapeutic, and diagnostic fields. Here, we present AptaNet—a new deep neural network—to predict the aptamer–protein interaction pairs by integrating features derived from both aptamers and the target proteins. Aptamers were encoded by using two different strategies, including k-mer and reverse complement k-mer frequency. Amino acid composition (AAC) and pseudo amino acid composition (PseAAC) were applied to represent target information using 24 physicochemical and conformational properties of the proteins. To handle the imbalance problem in the data, we applied a neighborhood cleaning algorithm. The predictor was constructed based on a deep neural network, and optimal features were selected using the random forest algorithm. As a result, 99.79% accuracy was achieved for the training dataset, and 91.38% accuracy was obtained for the testing dataset. AptaNet achieved high performance on our constructed aptamer-protein benchmark dataset. The results indicate that AptaNet can help identify novel aptamer–protein interacting pairs and build more-efficient insights into the relationship between aptamers and proteins. Our benchmark dataset and the source codes for AptaNet are available in: https://github.com/nedaemami/AptaNet.

show abstract

Section: Discussionmentioning

confidence: 99%

AptaNet as a deep learning approach for aptamer–protein interaction prediction

Emami

Ferdousi

2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This method could compensate for the limitations of the similarity-based methods. This approach can be categorized within the sequence and structural-based methods and consensus identifiers ( Xie et al, 2020 ). Moreover, the use of ML methods for the prediction of PPIs has represented improved performance in comparison to other conventional approaches, such as neural networks.…”

Section: In Silico Tools For the Prediction Of Protein–protein Interfacesmentioning

confidence: 99%

In silico Approaches for the Design and Optimization of Interfering Peptides Against Protein–Protein Interactions

Hashemi

Zarei

Fath

et al. 2021

Front. Mol. Biosci.

View full text Add to dashboard Cite

Large contact surfaces of protein–protein interactions (PPIs) remain to be an ongoing issue in the discovery and design of small molecule modulators. Peptides are intrinsically capable of exploring larger surfaces, stable, and bioavailable, and therefore bear a high therapeutic value in the treatment of various diseases, including cancer, infectious diseases, and neurodegenerative diseases. Given these promising properties, a long way has been covered in the field of targeting PPIs via peptide design strategies. In silico tools have recently become an inevitable approach for the design and optimization of these interfering peptides. Various algorithms have been developed to scrutinize the PPI interfaces. Moreover, different databases and software tools have been created to predict the peptide structures and their interactions with target protein complexes. High-throughput screening of large peptide libraries against PPIs; “hotspot” identification; structure-based and off-structure approaches of peptide design; 3D peptide modeling; peptide optimization strategies like cyclization; and peptide binding energy evaluation are among the capabilities of in silico tools. In the present study, the most recent advances in the field of in silico approaches for the design of interfering peptides against PPIs will be reviewed. The future perspective of the field and its advantages and limitations will also be pinpointed.

show abstract

“…Several scholars have also used Adam[36] to achieve an optimized cross-entropy loss detection function at a dropout rate of 0.5 [37]. The softmax function [38] function is seen as a class probability.…”

Section: Figure 2 High Correlation With Our Target Valuementioning

confidence: 99%

Pathway-Specific Protein Domains (PSPD) Discrimination by Using a Hybrid Feature Space Based on Deep Neural Network(DNN)

Ali¹,

Lei²,

Zhang³

et al. 2020

Preprint

View full text Add to dashboard Cite

The Pathway-specific protein domains (PSPDs) are important tools in examining drug growth as they provide a fast, reliable, and inexpensive way of estimating complex new molecular targets in specific diseases. The protein architecture prevents the formation of a direct correlation between signal transduction behavior and cellular structure. Accordingly, protein–tissue factor pathway inhibitor 2 isotypes 1 precursors have been used to encode peptide sequence information into specific feature structures. The measurable structure-activity classification model obtained by machine learning technology can predict pathway-specific protein interactions and new signaling peptides. We introduce deep neural network (DNN)-based PSPDs, abbreviated as DNNPSPDs, as the first pathway-specific protein domain that is built based on five extant models, namely, the AAindex, pseudo-amino acid composition, amino acid composition, composition mood of pseudoamino acids, and dipeptide composition. A total of 900 proteins with undetermined roles collected from the PDB data base are tested to evaluate the predictive power of this model. Various combinations of the available feature selection technologies are also combined to process a hybrid function space. DNNPSPDs predicts PSPDs by using features that are automatically learned from primary protein sequences. The sequences of pathway-associated proteins are sequentially fed into and decoded in neural network layers. Several classifications are also employed. DNNPSPDs achieves a prediction accuracy of 0.957 at a Matthew’s correlation coefficient (MCC) of 91.86%, with DPC, and 2nd achieve high prediction score 0.936 at Matthew’s correlation coefficient (MCC) of 88.02%, accuracy which is probably better. In terms of ROC–AUC, DNNPSPDs achieves a ROC–AUC curve of 0.982, which is larger than that of the other machine learning classifiers. A study using an alternative dataset reveals that our primary pathways, as pathway-specific protein domains, have accurate and reliable associations, thereby proving the viability of the proposed DNNPSPDs.

show abstract

Prediction of Protein–Protein Interaction Sites Using Convolutional Neural Network and Improved Data Sets

Cited by 49 publications

References 70 publications

AptaNet as a deep learning approach for aptamer–protein interaction prediction

AptaNet as a deep learning approach for aptamer–protein interaction prediction

In silico Approaches for the Design and Optimization of Interfering Peptides Against Protein–Protein Interactions

Pathway-Specific Protein Domains (PSPD) Discrimination by Using a Hybrid Feature Space Based on Deep Neural Network(DNN)

Contact Info

Product

Resources

About