Hum-mPLoc 3.0: prediction enhancement of human protein subcellular localization through modeling the hidden correlations of gene ontology and functional domain features

Zhou, Hang; Yang, Yang; Shen, Hong-Bin

doi:10.1093/bioinformatics/btw723

Cited by 94 publications

(70 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The subcellular localization of the identified proteins in this paper was analyzed according to the dataset from the Hum-mPLoc 3.0 [29][30][31]. Protein copy number per cell (pc/c) of the identified proteins was searched from a previous report by Wisniewski et al [32].…”

Section: Discussionmentioning

confidence: 99%

Design of five‐layer gold nanoparticles self‐assembled in a liquid open tubular column for ultrasensitive nano‐LC‐MS/MS proteomic analysis of 80 living cells

Shao

Zhang

2017

Proteomics

View full text Add to dashboard Cite

In this work, for the first time, a liquid open tubular column modified by five-layer gold nanoparticles and linked with C (GNPs@C ) was designed and fabricated for nano-LC-MS/MS analysis of 80 living cells. Sixty nanometer gold nanoparticles were self-assembled layer by layer on the inner wall of a 20 μm id fused-silica capillary. C was then linked on the gold nanoparticles to make the liquid open tubular column show hydrophobic character. Enough loading capacities for analysis of 80 living cells, ∼100 fmol for pk-10 and ∼30 fmol for insulin, were obtained with the 2 m × 20 μm id five-layer GNPs@C open tubular column. The open tubular column was used in an online pretreatment and direct nano-LC-MS/MS analysis system to analyze 80 living HepG2 cells. In total, 650 proteins were identified in triplicate runs. The subcellular localization of the identified proteins showed that our system had no bias toward different cellular compartments. Protein copy number per cell of the identified proteins showed that the detection limit could reach 50 zmol and the abundance of the identified proteins could cover a dynamic range of 6 orders.

show abstract

Section: Discussionmentioning

confidence: 99%

Design of five‐layer gold nanoparticles self‐assembled in a liquid open tubular column for ultrasensitive nano‐LC‐MS/MS proteomic analysis of 80 living cells

Shao

Zhang

2017

Proteomics

View full text Add to dashboard Cite

show abstract

“…To further demonstrate the performance of our method, we compare our BCMRFs with the only available PPI-based approach for predicting human protein SCLs, DC-kNN [17] and the state-of-art protein feature-based method Hum-mPLoc 3.0 [26]. DC-kNN is a physical PPI-based prediction method using a k-nearest neighbors classi cation with binary reference approach.…”

Section: Comparison With Existing Methodsmentioning

confidence: 99%

“…Hum-mPLoc 3.0 [26] is the state-of-the-art feature-based SCL predictor speci cally for human proteins. It predicts SCLs based on the amino acid sequence of proteins through modeling the hidden Table 4 demonstrate that our method achieves better performance.…”

Section: Comparison With Existing Methodsmentioning

confidence: 99%

“…• Hum-mPLoc 3.0 is a most recent protein feature-based SCL predictor for human proteins [26]. The predicted SCLs of 5390 human proteins from their database are used for the comparison.…”

Section: Comparison Partnersmentioning

confidence: 99%

“…(5) We performed experiments evaluating the performance on a human protein SCL benchmark set. Our method outperforms the state-of-the-art methods including the PPI-based approache DC-kNN [17] and the feature-based approach Hum-mPLoc 3.0 [26]. The rest of the paper is organized as follows: In section 2 we introduce our MRFs and the corresponding learning procedure.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Bayesian Collective Markov Random Fields for Subcellular Localization Prediction of Human Proteins

Zhu

Ester

2017

Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology,and Health Informatics

View full text Add to dashboard Cite

Advanced biotechnology makes it possible to access a multitude of heterogeneous proteomic, interactomic, genomic, and functional annotation data. One challenge in computational biology is to integrate these data to enable automated prediction of the Subcellular Localizations (SCL) of human proteins. For proteins that have multiple biological roles, their correct in silico assignment to di erent SCL can be considered as an imbalanced multi-label classi cation problem. In this study, we developed a Bayesian Collective Markov Random Fields (BCMRFs) model for multi-SCL prediction of human proteins. Given a set of unknown proteins and their corresponding protein-protein interaction (PPI) network, the SCLs of each protein can be inferred by the SCLs of its interacting partners. To do so, we integrate PPIs, the adjacency of SCLs and protein features, and perform transductive learning on the re-balanced dataset. Our experimental results show that the spatial adjacency of the SCLs improves multi-SCL prediction, especially for the SCLs with few annotated instances. Our approach outperforms the state-of-art PPIbased and feature-based multi-SCL prediction method for human proteins. KEYWORDSHuman protein subcellular localization; markov random eld; transductive learning; imbalanced multi-label classi cation.

show abstract

Tumor‐derived endomucin promotes colorectal cancer proliferation and metastasis

Huang

Sun

et al. 2022

Cancer Medicine

View full text Add to dashboard Cite

Background Endomucin (EMCN) is a type I transmembrane glycoprotein and a mucin‐like component of the endothelial cell glycocalyx. The mechanism of EMCN action in colorectal cancer (CRC) remains unclear. Aims Our aim was to explore the role of EMCN in the progression of CRC. Materials & Methods We examined EMCN expression in CRC tissues and normal para‐carcinoma tissues. The function and mechanisms of EMCN were checked in CRC cell lines and in mouse xenograft. Additionally, we used co‐immunoprecipitation and mass spectrometry to identify the potential EMCN‐binding proteins. Functional annotation analysis showed where these genes were enriched. Results We found that EMCN was overexpressed in tumor tissues compared with that in normal para‐carcinoma tissues. We also found that overexpression of EMCN induced CRC proliferation and metastasis both in vitro and in vivo. EMCN knockdown prevents epithelial‐mesenchymal transition in vitro. We identified 178 potential EMCN‐binding partners. Furthermore, functional annotation analysis indicated that these genes were considerably enriched in carcinogenic‐related functions and pathways. Collectively, the identification of EMCN‐binding partners enhanced our understanding of the mechanism of EMCN‐mediated malignant phenotypes, and this research may provide valuable insights into the molecular mechanisms underlying CRC. Conclusion Tumor‐derived endomucin promotes colorectal cancer proliferation and metastasis. We identified 178 EMCN‐binding proteins and initially screened three potential EMCN‐interacting proteins: NALCN, and TPM2, ANKK1. Our study provides valuable insights into the molecular mechanisms underlying CRC development.

show abstract

Hum-mPLoc 3.0: prediction enhancement of human protein subcellular localization through modeling the hidden correlations of gene ontology and functional domain features

Abstract: Supplementary data are available at Bioinformatics online.

Cited by 94 publications

References 50 publications

Design of five‐layer gold nanoparticles self‐assembled in a liquid open tubular column for ultrasensitive nano‐LC‐MS/MS proteomic analysis of 80 living cells

Design of five‐layer gold nanoparticles self‐assembled in a liquid open tubular column for ultrasensitive nano‐LC‐MS/MS proteomic analysis of 80 living cells

Bayesian Collective Markov Random Fields for Subcellular Localization Prediction of Human Proteins

Tumor‐derived endomucin promotes colorectal cancer proliferation and metastasis

Contact Info

Product

Resources

About