Subcellular locations of potential cell wall proteins in plants: predictors, databases and cross-referencing. (9th May 2017)

• Record Type: Journal Article
• Authors: Wu, Xiaolin; Zhang, Qinbin; Wu, Zhaokun; Tai, Fuju; Wang, Wei
• Is Part Of: Briefings in bioinformatics. Volume 19:Number 6(2018:Nov.)
• Page Start: 1130
• Page End: 1140
• Publication Date: 2017-05-09
• Subjects: bioinformatics prediction -- cell wall proteins -- plant cells -- subcellular location -- Web servers; Genetics -- Data processing -- Periodicals; Molecular biology -- Data processing -- Periodicals; Genomes -- Data processing -- Periodicals; 572.80285
• DOI: 10.1093/bib/bbx050
• Languages: English
• ISSNs: 1467-5463

Abstract

Abstract: The cell wall is the most striking feature that distinguishes plant cells from animal cells. It plays an essential role in cell shape, stability, growth and protection. Despite being present in small amounts, cell wall proteins (CWPs) are crucial components of the cell wall. The cell wall proteome generally consists of sensu stricto CWPs, apoplast proteins and extracellular secreted proteins. Currently, there is a need for the bioinformatics analysis of a tremendous number of protein sequences that have been generated from genomic, transcriptomic and proteomics research. Compared with intracellular proteins, the location prediction of CWPs is challenging because many aspects of these proteins have not been experimentally characterized, and there are no CWP-trained, specific predictors available. By introducing the biological relevance (particularly molecular aspects) of the cell wall and CWPs, we critically evaluated the accuracy of 16 state-of-the-art predictors and classical predictors for the prediction of CWPs using an independent database of Arabidopsis and rice proteins. All experimentally verified CWPs and non-CWPs were retrieved from the UniProt Knowledgebase. Based on the evaluation, we currently recommend the predictors mGOASVM, HybridGO-Loc and FUEL-mLoc for CWPs. Furthermore, we outlined the public databases that can be used to cross-reference the subcellular location of CWPs. We illustrate a flowchart of the subcellular location prediction and a cross-reference of possible CWPs. Finally, we discuss challenges and perspectives in the bioinformatics analysis of CWPs. It is hoped that this article will provide practical guidance regarding CWPs for nonspecialists and provide insight for bioinformatics experts to develop computational tools for CWPs.