Proteome-wide, Structure-Based Prediction of Protein-Protein Interactions/New Molecular Interactions Viewer

Abstract

Determining the complete Arabidopsis (Arabidopsis thaliana) protein-protein interaction network is essential for understanding the functional organization of the proteome. Numerous small-scale studies and a couple of large-scale ones have elucidated a fraction of the estimated 300,000 binary protein-protein interactions in Arabidopsis. In this study, we provide evidence that a docking algorithm has the ability to identify real interactions using both experimentally determined and predicted protein structures. We ranked 0.91 million interactions generated by all possible pairwise combinations of 1,346 predicted structure models from an Arabidopsis predicted "structure-ome" and found a significant enrichment of real interactions for the topranking predicted interactions, as shown by cosubcellular enrichment analysis and yeast two-hybrid validation. Our success rate for computationally predicted, structure-based interactions was 63% of the success rate for published interactions naively tested using the yeast two-hybrid system and 2.7 times better than for randomly picked pairs of proteins. This study provides another perspective in interactome exploration and biological network reconstruction using protein structural information. We have made these interactions freely accessible through an improved Arabidopsis Interactions Viewer and have created community tools for accessing these and ;2.8 million other protein-protein and protein-DNA interactions for hypothesis generation by researchers worldwide. The Arabidopsis Interactions Viewer is freely available at http://bar.utoronto.ca/interactions2/. Proteins rarely work alone, and most of the time they function in concert with other proteins or macromolecules. In Arabidopsis (Arabidopsis thaliana), the total number of binary interactions is estimated to be around 300,000 (Arabidopsis Interactome Mapping Consortium, 2011), but so far, only a small fraction of those interactions have been studied. Currently, there are 36,329 experimentally confirmed and 70,944 interolog-predicted protein-protein interactions (PPIs) in the Bio-Analytic Resource (BAR) interactions database (Geisler-Lee et al., 2007) that can be queried through the Arabidopsis Interactions Viewer (AIV). This huge gap indicates there is still a long way to go in elucidating the Arabidopsis interactome, both experimentally and computationally. With the arguable exception of the yeast two-hybrid method (Arabidopsis Interactome Mapping Consortium, 2011) or split ubiquitin method (Chen et al., 2012), traditional experimental methods for determining PPIs, such as mass spectrometry (Van Leene et al., 2007), protein microarrays (Popescu et al., 2007), and others (Zhang et al., 2010; Fukao, 2012), cannot readily be extended to determine the whole Arabidopsis interactome. Interolog-based computational PPI prediction methods (Geisler-Lee et al., 2007) can have a large-scale predictive ability but cannot

PlaPPISite: a comprehensive resource for plant protein-protein interaction sites

“…Regarding the interactome of A. thaliana, the number of experimental PPIs is relatively large, and some predicted interactomes have been publicly available, which has allowed us to directly compare our predicted PPIs with some existing interactomes. To this end, we collected 9065 predicted highly reliable A. thaliana PPIs (S-PPIs) from [43], which was based on a docking scoring algorithm using both experimentally determined and predicted protein structures. The self-interactions and interactions with proteins not appearing in our collected A. thaliana proteome were removed, and 8358 PPIs were finally retained.…”

“…To increase the coverage of these 13 plant interactomes, a large amount of predicted PPI data was collected in PlaP-PISite, although the reliability of predicted PPIs is always controversial. Even though three pieces of indirect evidence and a direct comparison with a predicted A. thaliana interactome developed by [43] have been provided to prove the acceptable reliability of the PPI prediction, the predicted PPIs in PlaPPISite may inevitably contain large volumes of false positives. Two efforts have been made to effectively guide users to use the predicted PPI data properly.…”

A host–pathogen interactome uncovers phytopathogenic strategies to manipulate plant ABA responses

“…This approach generated the ATIN consisting of 476 interactions between 97 Arabidopsis ABA-regulated proteins and 56 P. syringae T3SEs (Figure 1a). An interactive view of this network is available at http://bar.utoronto.ca/ATIN/ArabidopsisInter actome.html and in the Bio-Analtyic Resource's Arabidopsis Interactions Viewer 2 at http://bar.utoronto.ca/inte ractions2/ (Dong et al, 2019). A full list of all PPIs in ATIN are available in Table S1.…”

“…Interaction scores as described in Lumba et al . () for all 476 PPIs in ATIN are presented in Table S1 and can be visualized interactively at http://bar.utoronto.ca/ATIN/ArabidopsisInteractome.html, for the ATIN data set itself, and at http://bar.utoronto.ca/interactions2/ (Dong et al ., ) for the ATIN and ~80,000 other protein‐protein interactions. Data are also viewable in ePlant (Waese et al ., ).…”

Cytological and Proteomic Analysis of Wheat Pollen Abortion Induced by Chemical Hybridization Agent

“…Meanwhile, proteomics is a powerful approach to study the molecular processes of pollen development, and also a powerful complement to the whole genome sequencing [24]. Based on the proteins identified in this work, a detailed pathway was successfully constructed and protein-protein interactions were more clearly understood [25]. To date, there has been an increased application of proteomics approaches to study the pollen reproduction or pollen responses to abiotic stress, for example, Lycopersicon esculentum (Response to heat-stress and ethylene) [18], Oryza sativa (Response to high temperature-stress) [23], Triticum aestivum (Response to drought-stress) [26].…”