EuLoc: a web-server for accurately predict protein subcellular localization in eukaryotes by incorporating various features of sequence segments into the general form of Chou’s PseAAC

Abstract

The function of a protein is generally related to its subcellular localization. Therefore, knowing its subcellular localization is helpful in understanding its potential functions and roles in biological processes. This work develops a hybrid method for computationally predicting the subcellular localization of eukaryotic protein. The method is called EuLoc and incorporates the Hidden Markov Model (HMM) method, homology search approach and the support vector machines (SVM) method by fusing several new features into Chou's pseudo-amino acid composition. The proposed SVM module overcomes the shortcoming of the homology search approach in predicting the subcellular localization of a protein which only finds low-homologous or non-homologous sequences in a protein subcellular localization annotated database. The proposed HMM modules overcome the shortcoming of SVM in predicting subcellular localizations using few data on protein sequences. Several features of a protein sequence are considered, including the sequence-based features, the biological features derived from PROSITE, NLSdb and Pfam, the post-transcriptional modification features and others. The overall accuracy and location accuracy of EuLoc are 90.5 and 91.2 %, respectively, revealing a better predictive performance than obtained elsewhere. Although the amounts of data of the various subcellular location groups in benchmark dataset differ markedly, the accuracies of 12 subcellular localizations of EuLoc range from 82.5 to 100 %, indicating that this tool is much more balanced than other tools. EuLoc offers a high, balanced predictive power for each subcellular localization. EuLoc is now available on the web at http://euloc.mbc.nctu.edu.tw/.

Genome-wide analysis and expression profiling of DNA-binding with one zinc finger (Dof) transcription factor family in potato

“…StDof putative bipartite NLS showed two basic regions (B1 and B2) separated by a long spacer segment 17 amino acids long similar to the previously mapped bipartite NLS (Krebs et al, 2010). The protein subcellular localization prediction tool, EuLoc (Chang et al, 2013) showed that all StDof proteins are localized to the nucleus without exception.…”

“…Potatospecific StDof protein domains were searched against the PROSITE database (http://prosite.expasy.org/) for protein domains with known functions (Sigrist et al, 2013). Subcellular localization of the StDof proteins was predicted using the EuLoc protein localization prediction web server (Chang et al, 2013).…”

iSS-PseDNC: Identifying Splicing Sites Using Pseudo Dinucleotide Composition

“…Only the jackknife test is the least arbitrary that can always yield a unique result for a given benchmark dataset. Therefore, the jackknife test has been widely recognized and increasingly adopted by investigators to examine the quality of various predictors (see, e.g., [42, 58, 59, 62, 64, 66, 67, 70, 103–107]). Therefore, in this study, the jackknife test was also used to examine the performance of the predictor.…”

Progress in Machine Learning-based Predicting Subcellular Localizations of Proteins with Multiple Sites

“…With the aid of kernel function, a non-linear classification problem in a low dimension space can be transformed to a linear one after mapped it to a high dimension space. Recently, algorithms based on SVM are emerged massively [19,20].…”