Prediction of Protein-Protein Interactions with LSTM Deep Learning Model

Protein-protein interactions (PPI) has a vital role in molecular biology and bioinformatics since they are the key organisms which give information about cellular, its structure and its functions. In recent years many methods and techniques are proposed in order to perform PPI’s yet they are suffered from operational time, and large costs as well as low prediction accuracy. In this study, we performed a deep learning approach to resolve these problems. To do that we introduced a LSTM architecture to predict protein-protein interactions by applying both ProtVec and protein signatures methods. VCP (valosin-containing protein) which is associated with H. Pylori is considered in this work. The performance of the method determined by log-loss, ROC, and classification accuracy. The proposed method showed a good predictive ability yet there is still more works need to be performed to improve the results of PPI prediction studies with respect to deep learning and machine learning approaches.

[1]  Gary D Bader,et al.  Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry , 2002, Nature.

[2]  Gajendra P S Raghava,et al.  Classification of Nuclear Receptors Based on Amino Acid Composition and Dipeptide Composition* , 2004, Journal of Biological Chemistry.

[3]  Tatiana Tatusova,et al.  NCBI Reference Sequence (RefSeq): a curated non-redundant sequence database of genomes, transcripts and proteins , 2004, Nucleic Acids Res..

[4]  B. Séraphin,et al.  A generic protein purification method for protein complex characterization and proteome exploration , 1999, Nature Biotechnology.

[5]  E. Marcotte,et al.  A flaw in the typical evaluation scheme for pair-input computational predictions , 2012, Nature Methods.

[6]  Xiujun Gong,et al.  Deep Neural Network Based Predictions of Protein Interactions Using Primary Sequences , 2018, Molecules.

[7]  Luhua Lai,et al.  Sequence-based prediction of protein protein interaction using a deep-learning algorithm , 2017, BMC Bioinformatics.

[8]  Emmanuel D Levy,et al.  Evolution and dynamics of protein interactions and networks. , 2008, Current opinion in structural biology.

[9]  Yu Yao,et al.  DeepPPI: Boosting Prediction of Protein-Protein Interactions with Deep Neural Networks , 2017, J. Chem. Inf. Model..

[10]  Ehsaneddin Asgari,et al.  ProtVec: A Continuous Distributed Representation of Biological Sequences , 2015, ArXiv.

[11]  G. Frankel,et al.  Yeast two-hybrid system survey of interactions between LEE-encoded proteins of enteropathogenic Escherichia coli. , 2003, Microbiology.

[12]  Mansur R. Kabuka,et al.  Multimodal Deep Representation Learning for Protein-Protein Interaction Networks , 2018, 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[13]  SIGNET: A Neural Network Architecture for Predicting Protein-Protein Interactions , 2017 .

[14]  Jean-Loup Faulon,et al.  The Signature Molecular Descriptor. 2. Enumerating Molecules from Their Extended Valence Sequences , 2003, J. Chem. Inf. Comput. Sci..

[15]  Jean-Loup Faulon,et al.  Developing a methodology for an inverse quantitative structure-activity relationship using the signature molecular descriptor. , 2002, Journal of molecular graphics & modelling.

[16]  Lucian Ilie,et al.  SPRINT: ultrafast protein-protein interaction prediction of the entire human interactome , 2017, BMC Bioinformatics.

[17]  Jean-Loup Faulon,et al.  Predicting protein-protein interactions using signature products , 2005, Bioinform..

[18]  Long Zhang,et al.  Protein-protein interactions prediction based on ensemble deep neural networks , 2019, Neurocomputing.

[19]  Yuehui Chen,et al.  A novel method for prediction of protein interaction sites based on integrated RBF neural networks , 2012, Comput. Biol. Medicine.