idenPC-CAP: Identify protein complexes from weighted RNA-protein heterogeneous interaction networks using co-assemble partner relation

Abstract

Protein complexes play important roles in most cellular processes. The available genome-wide protein–protein interaction (PPI) data make it possible for computational methods identifying protein complexes from PPI networks. However, PPI datasets usually contain a large ratio of false positive noise. Moreover, different types of biomolecules in a living cell cooperate to form a union interaction network. Because previous computational methods focus only on PPIs ignoring other types of biomolecule interactions, their predicted protein complexes often contain many false positive proteins. In this study, we develop a novel computational method idenPC-CAP to identify protein complexes from the RNA-protein heterogeneous interaction network consisting of RNA–RNA interactions, RNA-protein interactions and PPIs. By considering interactions among proteins and RNAs, the new method reduces the ratio of false positive proteins in predicted protein complexes. The experimental results demonstrate that idenPC-CAP outperforms the other state-of-the-art methods in this field.

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

“…In our experiments, we primarily used an identity matrix as feature embedding. Moreover, we also tried a custom feature matrix, which was built using the RNA-RNA interaction networks and RNA-protein interaction networks [24]. For the node classification task, the adjacency matrix and feature matrix were required as input.…”

“…The results of this feature learning algorithm are provided in Table 2. Results for other methods are obtained from [24]. The number of epochs was 400, in this case, and a hidden layer sizes 512 were selected for this process with Adam as optimizer.…”

“…4) Approaches based on statistics, relying on the employment of statistical concepts for clustering proteins, e.g. how many shared neighbors pair of proteins have, and on notions of referential attachments for module members with other elements within the module; this includes SL [22], idenPC-MIIP [23], idenPC-CAP [24] and Farutin [1]. 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].…”

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

“…Approaches based on statistics, relying on the employment of statistical concepts for clustering proteins, e.g. how many shared neighbors pair of proteins have, and on notions of referential attachments for module members with other elements within the module; this includes SL [23], idenPC-MIIP [24], idenPC-CAP [25] and Farutin [2]. 5.…”

“…Apart from this, we have also tried a custom feature matrix. The RNA-RNA interaction networks and RNA-protein interaction networks [49] were used to build the feature matrix. For the node classification task, the adjacency matrix and feature matrix are required as input.…”