TESTLoc: protein localization prediction based on ESTs

Introduction
Download
Datasets and Supplementary Information
Useful links
Contact us

Introduction

TESTLoc is a subcellular localization prediction tool specifically tailored for Expressed Sequence Tags (ESTs). The current model predicts the plant proteins targeted to nine locations: cytosol, endoplasmic reticulum, extracellular space, mitochondria, nucleus,peroxisomes, plasma membrane, vacuole, and chloroplast. TESTLoc also allows users to train new models with their own data. It takes as input the EST-DNA sequences, and outputs the probability that the protein is transported to a given subcellular compartment.

TESTLoc is composed of two parts, EST ORF translation and localization prediction. The first part uses the tool Prot4EST. Alternatively, users may use as input already translated ESTs.


Download

TESTLoc is written in perl, and is designed for Linux system.

TESTLoc.tar.gz

Prot4EST and ESTScan are included in the package. Please visit the parlicular websites for installation instruction. Note that for using TESTLoc, LIBSVM 2.8 and BLAST have to be installed in the system and the corresponding executables must be included the environmental path. 


Datasets used for training and evaluation

Expanded plant data

Arabidopsis data

Patterns of tetra-peptides in our dataset


Useful Links

EST datbases:

NCBI dbEST

TbestdDB

The Gene Index Project


Contac Us

testloc@bch.umontreal.ca

Robert Cedergren Center for Bioinformatics and Genomics; Biochemistry Department, Université de Montréal, 2900 Edouard-Montpetit, Montreal, QC, H3T 1J4, Canada