A computational interactome and functional annotation for the human proteome

Garzón, José Ignacio; Deng, Lei; Murray, Diana; Shapira, Sagi; Petrey, Donald; Honig, Barry

doi:10.7554/elife.18715

Cited by 65 publications

(84 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This feature benefited from modeling with rice annotations and experimental data, allowing RicePPINet to discover PPIs specific for rice, which suggests that species‐specific modeling and optimization could effectively improve the accuracy and coverage of PPI prediction. In addition, we employed structural information in rice PPI prediction, which has been demonstrated to improve the reliability of predicted PPI networks by reducing false positives from a large number of non‐interacting protein pairs at the genome‐wide level (Zhang et al ., , ; Garzón et al ., ). All these factors contributed to the high predictability of RicePPINet for PPI discovery in rice.…”

Section: Discussionmentioning

confidence: 97%

“…A critical advantage of structure‐based approaches is their ability to identify the putative conserved interface, providing more information than any non‐structure‐based methods. Structure‐based methods have been demonstrated to be powerful for PPI prediction in model organisms, including Arabidopsis (Zhang et al ., , ; Garzón et al ., ). The availability of structure‐based PPI network models of Arabidopsis should be useful for other plants such as rice, as many processes are conserved between the two species; however, some rice gene networks associated with specific phenotypes are different to that in Arabidopsis because of high levels of evolutionary diversity between these two model plants (Devos et al ., ).…”

Section: Introductionmentioning

confidence: 97%

See 1 more Smart Citation

A computational interactome for prioritizing genes associated with complex agronomic traits in rice (Oryza sativa)

et al. 2017

View full text Add to dashboard Cite

Rice (Oryza sativa) is one of the most important staple foods for more than half of the global population. Many rice traits are quantitative, complex and controlled by multiple interacting genes. Thus, a full understanding of genetic relationships will be critical to systematically identify genes controlling agronomic traits. We developed a genome-wide rice protein-protein interaction network (RicePPINet, http://netbio.sjtu.edu.cn/riceppinet) using machine learning with structural relationship and functional information. RicePPINet contained 708 819 predicted interactions for 16 895 non-transposable element related proteins. The power of the network for discovering novel protein interactions was demonstrated through comparison with other publicly available protein-protein interaction (PPI) prediction methods, and by experimentally determined PPI data sets. Furthermore, global analysis of domain-mediated interactions revealed RicePPINet accurately reflects PPIs at the domain level. Our studies showed the efficiency of the RicePPINet-based method in prioritizing candidate genes involved in complex agronomic traits, such as disease resistance and drought tolerance, was approximately 2-11 times better than random prediction. RicePPINet provides an expanded landscape of computational interactome for the genetic dissection of agronomically important traits in rice.

show abstract

Section: Discussionmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 97%

A computational interactome for prioritizing genes associated with complex agronomic traits in rice (Oryza sativa)

et al. 2017

View full text Add to dashboard Cite

show abstract

“…They have also been used to predict PPIs for previously uncharacterized proteins, e.g. PrePPI [22][23][24]. Whereas it can be expected that these systematic PPI mapping projects will reach 2/32 saturation in the next few years, AP-MS and related approaches are fundamentally limited in their ability to detect compositional or abundance changes across multiple cell states and to detect concurrent changes in different complexes within the same sample.…”

Section: Introductionmentioning

confidence: 99%

SECAT: Quantifying differential protein-protein interaction states by network-centric analysis

Rosenberger

Heusel

Bludau

et al. 2019

Preprint

View full text Add to dashboard Cite

Protein-protein interactions (PPIs) play a critical role in virtually all cellular processes. Their context-dependent characterization is thus a key objective of proteomic research. We and others have previously shown that chromatographic fractionation of native protein complexes (e.g. through size-exclusion chromatography, SEC) can be effectively combined with high-throughput, bottom-up mass-spectrometry-based proteomics (e.g. data-independent acquisition-based SWATH-MS), to support proteome-wide characterization of protein complexes.To enable qualitative and quantitative comparison of the proteome organization encoded in these datasets, across multiple experimental conditions, scalable and robust 1/32 analysis strategies are required. To address this need, we developed the Size-Exclusion Chromatography Algorithmic Toolkit (SECAT), a novel network-centric strategy for the quantitation of protein complex profiles. SECAT elucidates proteins and their context-specific PPIs in terms of both abundance and connectivity. We validate algorithm predictions using publicly available datasets and compare them to established strategies to demonstrate that SECAT represents a more scalable and effective methodology to assess protein-network state, obviating the need to infer individual protein complexes. Further, by comparing PPI-networks in interphase and mitotic HeLa cells, we demonstrate SECAT's ability to provide novel insight about context-specific molecular mechanisms that differentiate cellular states.

show abstract

“…Proteins binding with RNA through specific residues have a profound effect on many biological processes such as protein synthesis [1], post-transcriptional modifications, and regulation of gene expression [2–4]. Determining these protein-RNA binding residues can help to elucidate the underlying mechanisms, to control biological processes, or to design RNA-based drug.…”

Section: Introductionmentioning

confidence: 99%

“…A series of machine learning methods [6] such as Naive Bayes, support vector machine (SVM), and random forest (RF), combined with amino acid sequence or protein three-dimensional structural characteristics [4, 7], have been proposed to identify RNA-binding residues. Jeong et al [8] build a neural network classifier to predict RNA-binding residues based on protein sequence and structural information.…”

Section: Introductionmentioning

confidence: 99%

A boosting approach for prediction of protein-RNA binding residues

et al. 2017

Self Cite

View full text Add to dashboard Cite

BackgroundRNA binding proteins play important roles in post-transcriptional RNA processing and transcriptional regulation. Distinguishing the RNA-binding residues in proteins is crucial for understanding how protein and RNA recognize each other and function together as a complex.ResultsWe propose PredRBR, an effectively computational approach to predict RNA-binding residues. PredRBR is built with gradient tree boosting and an optimal feature set selected from a large number of sequence and structure characteristics and two categories of structural neighborhood properties. In cross-validation experiments on the RBP170 data set show that PredRBR achieves an overall accuracy of 0.84, a sensitivity of 0.85, MCC of 0.55 and AUC of 0.92, which are significantly better than that of other widely used machine learning algorithms such as Support Vector Machine, Random Forest, and Adaboost. We further calculate the feature importance of different feature categories and find that structural neighborhood characteristics are critical in the recognization of RNA binding residues. Also, PredRBR yields significantly better prediction accuracy on an independent test set (RBP101) in comparison with other state-of-the-art methods.ConclusionsThe superior performance over existing RNA-binding residue prediction methods indicates the importance of the gradient tree boosting algorithm combined with the optimal selected features.

show abstract

A computational interactome and functional annotation for the human proteome

Cited by 65 publications

References 77 publications

A computational interactome for prioritizing genes associated with complex agronomic traits in rice (Oryza sativa)

A computational interactome for prioritizing genes associated with complex agronomic traits in rice (Oryza sativa)

SECAT: Quantifying differential protein-protein interaction states by network-centric analysis

A boosting approach for prediction of protein-RNA binding residues

Contact Info

Product

Resources

About