Higher Edu - Research dev card
Development from the higher education and research community
  • Creation or important update: 16/09/10
  • Minor correction: 16/09/10

HECTAR : subcellular localisation, heterokonts, support vector machine, metaclassifier, motif, annotation

This software was developed (or is under development) within the higher education and research community. Its stability can vary (see fields below) and its working state is not guaranteed.


General software features

HECTAR is the first published prediction method to detect N-terminal target peptides for heterokonts. Three main features characterise HECTAR:
The first is the hierarchical architecture of our method which enables the identification of the complex chloroplast bi-peptide address signal.
A second feature is the modularity of our method. Each of the three HECTAR modules is dedicated to identify one specific N-terminal target peptide. It is possible to analyse only a protein sequence with a specific module or a combination of modules. In this way, a larger variety of eukaryotic organisms can be analysed, as it is the case with the variants HECTARsec and HECTARmetfun.
The third feature which outlines our method is the use of metaclassifiers for the prediction process. In each HECTAR module, significant outputs of dedicated N-terminal target prediction methods are combined by a support vector machine. It has been shown that this kind of approach improves the quality of the prediction.

Context in which the software is used

In eukaryotes, most of the proteins that function in the organelles are encoded by the cell nucleus. These proteins are targeted to organelles by various address signals. Addressing proteins to complex plastids of heterokonts is enabled by bi-peptide signals and a well conserved motif (ASAFAP) at the cleavage site of the N-terminal signal peptide. HECTAR is able to detect these complex targeting signals and it can predict the five different categories of N-terminal targeting. A hierarchical approach consisting of several decision modules was designed to take into account protein transfer into complex chloroplasts of heterokonts. In each module prediction methods dedicated to the detection of a specific address signal are combined by a support vector machine.

In addition to HECTAR we have also created two variants of the method to allow subcellular localisation annotations of other organisms than heterokonts: the variant HECTARmetfun can be applied to identify secretory pathway address signals and mitochondrial transit peptides for metazoan and fungi nuclear-encoded proteins. The second variant, called HECTARsec, can detect N-terminal signal peptides and signal anchors for eukaryotic nuclear-encoded proteins.

HECTAR and its variants have been shown to have a high prediction performance.

Publications related to the software

HECTAR: a method to predict subcellular targeting in heterokonts. Gschloessl B, Guermeur Y, Cock JM., BMC Bioinformatics. 2008 Sep 23;9:393.

Development of a method which predicts N-Terminal target peptides and study of protein sorting in eukaryote genomes. Gschloessl B., PhD thesis, 2008, University Pierre and Marie Curie (Paris VI), Roscoff Biological Station, FRANCE