Introduction Methods Genomes Prediction

Update (05/28/2008): Fixed bugs related to move to new website of rostlab and improved query handling. Please report any bugs you encounter. Your patience during our move to a new website is appreciated. Description of predictions by placing mouse over question marks on output page.

Name your request (optional):
Results via Browser (default) or Email if provided:
Type of organism for which localization is to be predicted:
animal
plant
prokaryote
Type of protein sequence for which localization is to be predicted (non-membrane proteins only!) :
Full length protein sequence
Fragment of protein (more than 20 amino acid residues)
Note: Upto 100 protein sequences can be submitted at a time. If more than 10 sequences are submitted the job is run in low priority mode. The first line should be a protein identifier line and should begin with a '>' sign followed by the protein id. The protein id should have some letters of the alphabet and cannot be purely numerical (see below for further explanation of fasta sequence format and allowed protein id's). No prediction is returned for fragments shorter than 20 residues, since results are likely inaccurate.

Submit a Fasta file, with one or more protein sequences (non-membrane proteins only) :
Or paste:

If you found LOCtree useful, please cite: R Nair and B Rost,  Mimicking cellular sorting improves prediction of Subcellular Localization. Journal of Molecular Biology 2005 Apr 22;348(1):85-100.

The input must be in multiple fasta format.    All sequences must have unique IDs. The entire header line is taken as the ID. The ID cannot be solely numerical (>34 not allowed while >p34 is allowed). The sequence may contain the 20 amino acid characters plus B, Z, X, space, and newline in upper or lower case.

LOCtree was developed by Rajesh Nair. For help and suggestions, please contact nair@cubic.bioc.columbia.edu.