explanations in a web-based tool for high-throughput proteome annotations

PA-GOSUB (Proteome Analyst: GO Molecular Function and Subcellular Localization) is a publiclyavailable, web-based, searchable, and downloadable database that contains the sequences, predicted GO molecular functions, and predicted subcellular localizations of the more than 107,000 proteins from 10 model organisms (and growing), covering the major kingdoms and phyla for which annotated proteomes exist (http://www.cs.ualberta.ca/~bioinfo/PA/GOSUB). The PA-GOSUB database effectively expands the coverage of subcellular localization and GO function annotations by a significant factor (already over 5 for subcellular localization, as compared to Swiss-Prot v42.7) and more model organisms are being added to PA-GOSUB as their sequenced proteomes become available.

doi:10.1093/bioinformatics/bti372

Abstract — The subcellular locations of proteins are important functional annotations. An effective and reliable subcellular localization method is necessary for proteomics research. This paper introduces a new method—PairProSVM—to automatically predict the subcellular locations of proteins. The profiles of all protein sequences in the training set are constructed by PSI-BLAST and the pairwise profile-alignment scores are used to form feature vectors for training a support vector machine (SVM) classifier. It was found that PairProSVM outperforms the methods that are based on sequence alignment and amino-acid compositions even if most of the homologous sequences have been removed. PairProSVM was evaluated on Huang and Li’s and Gardy et al.’s protein datasets. The overall accuracies on these datasets reach 75.3 % and 91.9%, respectively, which are higher than or comparable to those obtained by sequence alignment and composition-based methods. Index Terms — Protein subcellular localization; sequence alignment; profile alignment; kernel methods; support vector machines. I.

for bacteria

This paper presents a system of recurrent neural networks which demonstrate an ability to detect residues belonging to specific targeting peptides with greater accuracy than current feed forward models. The system can subsequently be used for determining sub-cellular localisation of proteins and for understanding the factors underlying translocation. The work can be seen as building upon the currently popular series of predictors SignalP and TargetP, by exploiting the inherent bias for sequential pattern recognition exhibited by recurrent networks

Abstract. A new kernel has been developed for vectors derived from a coding scheme of the tri-peptide composition for protein sequences. This kernel defines the sequence similarity through a mapping that transforms a tri-peptide coding vector into a new vector based on a matrix formed by the high BLOSUM scores associated with pairs of tri-peptides. In conjunction with the use of support vector machines, the effectiveness of the new kernel is evaluated against the conventional coding schemes of k-peptide (k ≤ 3) for the prediction of subcellular localizations of proteins in Gram-negative bacteria. It is demonstrated that the new method outperforms all the other methods in a 5-fold cross-validation. Keywords: protein subcellular localization, Gram-negative bacteria, BLOSUM matrix, kernel, support vector machine. 1

Motivation: In general, each cell signaling pathway involves many proteins, each with one or more specific roles. As they are essential components of cell activity, it is important to understand how these proteins work – and in particular, to determine which of a species ’ proteins participate in each role. Experimentally determining this mapping of proteins to roles is difficult and time consuming. Fortunately, many pathways are similar across species, so we may be able to use known pathway information of one species to understand the corresponding pathway of another. Results: We present an automatic approach, Predict Signaling Pathway (PSP), that uses the signaling pathways in well-studied species to predict the roles of proteins in less-studied species. We use a machine learning approach to create a predictor that achieves a generalization F-measure of 78.2 % when applied to 11 different pathways across 14 different species. We also show our approach is very effective in predicting the pathways that have not yet been experimentally studied completely. Contact:

Nuclear localization of proteins is a crucial element in the dynamic life of the cell. It is complicated by the massive diversity of targeting signals and the existence of proteins that shuttle between the nucleus and cytoplasm. In spite of this, the majority of subcellular localization tools that predict nuclear proteins have been developed without involving dual localized proteins in the data sets. Hence, in general, the existing models are focused on predicting statically nuclear proteins, rather than nuclear localization itself. We present an independent analysis of existing nuclear localization predictors, using a non-redundant data set extracted from Swiss-Prot R50.0. We demonstrate that accuracy on truly novel proteins is lower than the previous estimations, and that existing models generalize poorly to dual localized proteins. We develop a model trained to identify nuclear proteins including dual localized proteins. The results suggest that using more recent data and including dual localized proteins improves the overall prediction. The final predictor Nucleo operates with a realistic success rate of 0.70 and a correlation coefficient of 0.38, as established on the independent test set.

subcellular localization and insights gained from comparative proteome analysis