Identification of Relevant Protein Interactions with Partial Knowledge: A Complex Network and Deep Learning Approach

Abstract

Protein–protein interactions (PPIs) are the basis for understanding most cellular events in biological systems. Several experimental methods, e.g., biochemical, molecular, and genetic methods, have been used to identify protein–protein associations. However, some of them, such as mass spectrometry, are time-consuming and expensive. Machine learning (ML) techniques have been widely used to characterize PPIs, increasing the number of proteins analyzed simultaneously and optimizing time and resources for identifying and predicting protein–protein functional linkages. Previous ML approaches have focused on well-known networks or specific targets but not on identifying relevant proteins with partial or null knowledge of the interaction networks. The proposed approach aims to generate a relevant protein sequence based on bidirectional Long-Short Term Memory (LSTM) with partial knowledge of interactions. The general framework comprises conducting a scale-free and fractal complex network analysis. The outcome of these analyses is then used to fine-tune the fractal method for the vital protein extraction of PPI networks. The results show that several PPI networks are self-similar or fractal, but that both features cannot coexist. The generated protein sequences (by the bidirectional LSTM) also contain an average of 39.5% of proteins in the original sequence. The average length of the generated sequences was 17% of the original one. Finally, 95% of the generated sequences were true.

Efforts to Minimise the Bacterial Genome as a Free-Living Growing System

“…Machine learning is a computational approach to apprehending representative patterns from a dataset of unknown structures. It has been used in various fields of biology, e.g., genetic prediction, protein engineering, and metabolic engineering [ 89 , 90 , 91 , 92 , 93 , 94 , 95 ], as it benefits biological studies from a system point of view. The advantage of machine learning-assisted medium optimization is the ability to predict the appropriate concentrations of medium components independent of personal professional experience or prior literature knowledge, according to a given dataset acquired experimentally.…”

Bioinformatics Applications in Chronic Diseases: A Comprehensive Review of Genomic, Transcriptomics, Proteomic, Metabolomics, and Machine Learning Approaches

“…These networks are built by detecting the physical connections between proteins, which can be accomplished experimentally using methods like yeast two-hybrid experiments, co-immunoprecipitation, and affinity purification combined with MS. Researchers can create detailed diagrams that depict the connection of proteins implicated in disease-related processes by mapping these interactions (Ortiz-Vilchis et al, 2023). The identification of important nodes or hubs is a vital feature of protein interaction networks in disease processes.…”

Recent Advances in Deep Learning for Protein-Protein Interaction Analysis: A Comprehensive Review

“…For PPI network prediction, Mahdipour et al [ 84 ] introduced RENA, an innovative method for PPI network alignment based on recurrent neural networks. Ortiz-Vilchis et al [ 85 ] employed a bidirectional LSTM model for generating relevant protein sequences with partial knowledge of interactions, demonstrating an ability to retain a significant portion of proteins in the original sequence.…”