A seed-extended algorithm for detecting protein complexes based on density and modularity with topological structure and GO annotations

Wang, Rongquan; Wang, Caixia; Sun, Liankun

doi:10.1186/s12864-019-5956-y

Cited by 13 publications

(37 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6. Approaches based on modularity, topological structure, overlapping information, and GO annotations, including CFinder [28] and [29], ClusterONE [30], and SE-DMTG [31].…”

Section: Local Search Approaches Based On Cost Focusing On the Extraction Of Modules From Interaction Graphsmentioning

confidence: 99%

Identifying Protein Complexes in Protein-protein Interaction Data using Graph Convolution Network

Zaki

Singh

2021

Preprint

View full text Add to dashboard Cite

Protein complexes are groups of two or more polypeptide chains that join together to build noncovalent networks of protein interactions. A number of means of computing the ways in which protein complexes and their members can be identified from these interaction networks have been created. While most of the existing methods identify protein complexes from the protein-protein interaction networks (PPIs) at a fairly decent level, the applicability of advanced graph network methods has not yet been adequately investigated. In this paper, we proposed various graph convolutional networks (GCNs) methods to improve the detection of the protein functional complexes. We first formulated the protein complex detection problem as a node classification problem. Second, the Neural Overlapping Community Detection (NOCD) model was applied to cluster the nodes (proteins) using a complex affiliation matrix. A representation learning approach, which combines the multi-class GCN feature extractor (to obtain the features of the nodes) and the mean shift clustering algorithm (to perform clustering), is also presented. We have also improved the efficiency of the multi-class GCN network to reduce space and time complexities by converting the dense-dense matrix operations into dense-spares or sparse-sparse matrix operations. This proposed solution significantly improves the scalability of the existing GCN network. Finally, we apply clustering aggregation to find the best protein complexes. A grid search was performed on various detected complexes obtained by applying three well-known protein detection methods namely ClusterONE, CMC, and PEWCC with the help of the Meta-Clustering Algorithm (MCLA) and Hybrid Bipartite Graph Formulation (HBGF) algorithm. The proposed GCN-based methods were tested on various publicly available datasets and provided significantly better performance than the previous state-of-the-art methods. The code and data used in this study are available from https://github.com/Analystharsh/GCN_complex_detection

show abstract

“…6. Approaches based on modularity, topological structure, overlapping information, and GO annotations, including CFinder [28] and [29], ClusterONE [30], and SE-DMTG [31].…”

Section: Local Search Approaches Based On Cost Focusing On the Extraction Of Modules From Interaction Graphsmentioning

confidence: 99%

Identifying Protein Complexes in Protein-protein Interaction Data using Graph Convolution Network

Zaki

Singh

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Gene Ontology (GO) annotations of proteins are classi ed mainly among three categories: biological process, cellular component, and molecular function [25]. GO annotations of the proteomics results were from the UniProt-Gene Ontology Annotation (GOA) database [26].…”

Section: Bioinformatics Analysesmentioning

confidence: 99%

Systematic Analysis of the Lysine Malonylome in Sanghuangporus Sanghuang

Wang

Zhang

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: Sanghuangporus sanghuang is a well-known traditional medicinal mushroom associated with mulberry, which has enormous economic and medical value. Despite being known for many years, proteomics studies of S. sanghuang have not been reported. In this paper, a series of technologies, such as high-performance liquid chromatography (HPLC) classification technology, malonyl peptide segment enrichment technology and proteomics technology, are combined to study the qualitative analysis of malonylation in S. sanghuang.Results: Strategies for qualitative proteomics included malonyl enrichment and high-resolution liquid chromatography-tandem mass spectrometry (LC-MS/MS). In total, 714 malonyl modification sites were matched to 255 different proteins. The analysis indicated that malonyl modifications are involved in a wide range of acellular functions and displayed a distinct subcellular localization.Bioinformatics analysis indicates that malonylated proteins are engaged in different metabolic pathways, including glyoxylate and dicarboxylate metabolism, glycolysis/gluconeogenesis, and the tricarboxylic acid (TCA) cycle; consequently, a total of 26 enzymes related to triterpene and polysaccharide biosynthesis were found to be malonylated.Conclusions: These findings suggest that malonylation is associated with many metabolic pathways, particularly the metabolism of bioactive compounds. This paper provides the first comprehensive survey of malonylation in S. sanghuang.

show abstract

“…5) Stochastic search methods based on population employed to develop algorithms used to detect communities and networks including CGA [2], IGA [6], and EHO-MCL [25]. 6) Approaches based on modularity, topological structure, overlapping information, and GO annotations, including CFinder [26] and [27], ClusterONE [28], and SE-DMTG [29]. 7) Graph-based clustering methods, which includes statistical-based measures methods such as [1] which uses the concept of statistical significance to measure the strength of the relationship between two nodes (proteins), which requires prior estimation of the pvalue.…”

Section: Introductionmentioning

confidence: 99%

Identifying Protein Complexes in Protein-Protein Interaction Data Using Graph Convolutional Network

Zaki¹,

Singh

Mohamed

2021

IEEE Access

View full text Add to dashboard Cite

Protein complexes are groups of two or more polypeptide chains that bind to form noncovalent networks of protein interactions. Over the past decade, researchers have created a number of means of computing the ways in which protein complexes and their members can be identified through these interaction networks. Although most of the existing methods identify protein functional complexes from the protein-protein interaction networks (PPIs) at a fairly decent level, the applicability of advanced graph network methods has not yet been adequately investigated. This paper proposes various graph convolutional network (GCN) methods to improve the detection of protein complexes. We first formulate the protein complex detection problem as a node classification problem. Then, we developed a Neural Overlapping Community Detection (NOCD) model to cluster the nodes (proteins) using a complex affiliation matrix. A representation learning approach, that combines a multi-class GCN feature extractor (to obtain the nodes' features) and a mean shift clustering algorithm (to perform the clustering), is also utilized. We convert the dense-dense matrix operations into dense-sparse or sparse-sparse matrix operations to improve the efficiency of the multi-class GCN network by reducing space and time complexities. The proposed solution significantly improves the scalability of the existing GCN. Finally, we apply clustering aggregation to find the best protein complexes. A grid search is then performed on various detected complexes obtained via three well-known protein detection methods, namely ClusterONE, CMC, and PEWCC, with the help of the Meta-Clustering Algorithm (MCLA) and the Hybrid Bipartite Graph Formulation (HBGF). We test the proposed GCN-based methods on various publicly available datasets and find that they perform significantly better than previous state-of-the-art methods. The code/data are available for free download from https://github.com/Analystharsh/GCN_complex_detection.

show abstract

A seed-extended algorithm for detecting protein complexes based on density and modularity with topological structure and GO annotations

Cited by 13 publications

References 79 publications

Identifying Protein Complexes in Protein-protein Interaction Data using Graph Convolution Network

Identifying Protein Complexes in Protein-protein Interaction Data using Graph Convolution Network

Systematic Analysis of the Lysine Malonylome in Sanghuangporus Sanghuang

Identifying Protein Complexes in Protein-Protein Interaction Data Using Graph Convolutional Network

Contact Info

Product

Resources

About