FCTP-WSRC: Protein–Protein Interactions Prediction via Weighted Sparse Representation Based Classification

Abstract

The task of predicting protein-protein interactions (PPIs) has been essential in the context of understanding biological processes. This paper proposes a novel computational model namely FCTP-WSRC to predict PPIs effectively. Initially, combinations of the F-vector, composition (C) and transition (T) are used to map each protein sequence onto numeric feature vectors. Afterwards, an effective feature extraction method PCA (principal component analysis) is employed to reconstruct the most discriminative feature subspaces, which is subsequently used as input in weighted sparse representation based classification (WSRC) for prediction. The FCTP-WSRC model achieves accuracies of 96.67%, 99.82%, and 98.09% for H. pylori, Human and Yeast datasets respectively. Furthermore, the FCTP-WSRC model performs well when predicting three significant PPIs networks: the single-core network (CD9), the multiple-core network (Ras-Raf-Mek-Erk-Elk-Srf pathway), and the cross-connection network (Wnt-related Network). Consequently, the promising results show that the proposed method can be a powerful tool for PPIs prediction with excellent performance and less time.

Assessment of community efforts to advance computational prediction of protein-protein interactions

“…In this project, although we tested many algorithms covering different categories, the classifiers based on protein sequence, for example, SVM [38][39][40][41] , RF 42,43 , FCTP 44 , and DPPI 45 , have not been tested for two reasons. First, those methods need to define a feature space for each link, which will lead to significant time complexity and memory requirement for HuRI (which has ~ 35 million unmapped PPIs).…”

Protein–protein interaction prediction with deep learning: A comprehensive review

“… Finally, protein binding affinity data were obtained from the SKEMPI dataset [320] , comprising 3,047 binding affinity changes after mutation of protein subunits within a protein complex for use in the affinity estimation task. ACT-SVM [2020] [289] Following [321] a nonredundant dataset including H. pylori and human PPI was created. The H. pylori dataset comprised 1,458 interacting and 1,457 noninteracting protein pairs, while the human dataset comprised 3,899 interacting and 4,262 noninteracting protein pairs.…”

“… BiLSTM-RF [2021] [291] A nonredundant human dataset was retrieved from the DIP database. Sequences were clustered using the CD-HIT tool based on sequence similarity to remove redundancy and establish a nonredundant human PPI dataset [325] , [321] . This dataset included 4,262 interacting protein pairs and 3,899 noninteracting protein pairs.…”

BoT-Net: a lightweight bag of tricks-based neural network for efficient LncRNA–miRNA interaction prediction

“…Whereas, the second paradigm combines lncRNA and miRNA sequences to formulate lncRNA-miRNA sequence pairs where every pair is treated as a single instance. These pairs are passed to a single head neural network which extracts important features before passing forward to the final classification layer [35].…”