Prediction of Protein Subcellular Localization
by Support Vector Machine
-- Sujun HUA --
-- Institute of Bioinformatics, Tsinghua Univ.--
SubLoc is a prediction system for protein subcellular localization based on amino acid composition alone.
Support Vector Machine (SVM) method is used in SubLoc.
SubLoc predicts locations of prokaryotic and eukaryotic sequences separately.
For prokaryotic sequences, SubLoc predicts three locations (cytoplasmic, periplasmic and extracellular).
For eukaryotic sequences, SubLoc predicts four locations (cytoplasmic, extracellular, mitochondrial and nuclear).
SubLoc v1.0 is based on the dataset of Reinhardt and Hubbard (Nucleic Acid Research, 1998, 26:2230-2236).
Sujun Hua & Zhirong Sun:
" Support Vector Machine Approach for Protein Subcellular Localization Prediction" , Bioinformatics, in press.
Predicting your protein sequences by SubLoc v1.0
Prokaryotic sequences (batch)
Eukaryotic sequences (batch)
Checking whether the protein is a membrane protein
SubLoc v1.0 does not predict the membrane protein and we recommend you to predict whether your protein is a membrane protein using the WWW prediction servers:
If you have any comments and problems, please contact Sujun Hua (firstname.lastname@example.org)
Last Modified: Mar. 20, 2001.