Deep ensemble model for sequence-based prediction of PPI: Self improved optimization assisted intelligent model

PPIs play a significant function in many biological processes. In many different areas, DL algorithms have delivered excellent results, but PPI prediction is one where they fall short. To offer a sequence-based prediction of PPI, this work employs a deep ensemble model. In the beginning, traits incl...

Full description

Saved in:
Bibliographic Details
Published in:Multimedia tools and applications 2024-03, Vol.83 (26), p.68135-68154
Main Authors: Srivastava, Deepak, Mall, Shachi, Singh, Suryabhan Pratap, Bhatt, Ashutosh, Kumar, Shailesh, Dheresh Soni
Format: Article
Language:English
Subjects:
Algorithms
Artificial neural networks
Belief networks
Biological activity
Biological effects
Computer Communication Networks
Computer Science
Data Structures and Information Theory
Information technology
Mean square errors
Methods
Multimedia
Multimedia Information Systems
Neural networks
Ontology
Optimization
Optimization algorithms
Prediction models
Proteins
Recurrent neural networks
Semantics
Signal transduction
Special Purpose and Application-Based Systems
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:PPIs play a significant function in many biological processes. In many different areas, DL algorithms have delivered excellent results, but PPI prediction is one where they fall short. To offer a sequence-based prediction of PPI, this work employs a deep ensemble model. In the beginning, traits including "enhanced semantic similarity, features based on gene ontologies, and sequence-based features" are extracted. A deep ensemble model is introduced that combines models like Deep Convolutional Neural Network (DCNN), Recurrent Neural Network (RNN), Deep Max out (DMO), and Deep Belief Network (DBN)" is then used to predict the outcomes of the retrieved features. To improve the prediction model, the training is done by the Chaotic Initialized COOT Optimization Algorithm (CI-COA) by optimizing the training weights of DCNN. The performance of the chosen strategy is finally shown through several measures.
ISSN:1573-7721
1380-7501
1573-7721
DOI:10.1007/s11042-024-18738-3