TM-Finder Help
About the TM-Finder
The TM-Finder is a Trans-Membrane protein prediction tool that was developed based on an algorithm, data, and prototype program produced by Dr. Charles Deber and his research team. The program shows the regions predicted to be trans-membrane based on the helicity and hydrophobicity of the neighbouring amino acid sequences.
A TM segment is predicted if both the average hydrophobicity and helicity values of a core of amino acids is greater than the threshold (0.4 for hydrophobicity and 1.1 for helicity). The average hydrophobicity and helicity are based on the average hydrophobicity and helicity of each amino acid along with a specified number of amino acids on its N-terminal (amino terminal) and C-terminal (carboxy terminal) end. The number of amino acids to be taken from the N-terminal and C-terminal end can be specified by choosing a number in the "N-terminal window" and "C-terminal window" boxes. Other parameters used to calculate the TM-segments are the Minimum core length (which specifies the minimum number of amino acids whose average hydrophobicity and helicity should be greater than the threshold to be thought of as a potential TM-segment core), Gap-length (which specifies the number of amino acids that could be added to the TM-segment core if they are surrounded by TM cores on both sides) and Minimum segment length (minimum number of amino acids having average hydrophobicity and helicity greater than threshold to form a TM-segment).
Input to the Program
In order to run, the program must be given a sequence of amino acids. The program can also accept amino acid sequences in FASTA format which has a header beginning with ">", for example,
>gi|532319|pir|TVFV2E|TVFV2E envelope protein
ELRLRYCAPAGFALLKCNDADYDGFKTNCSNVSVVHCTNLMNTTVTTGLLLNGSYSENRT
QIWQKHRTSNDSALILLNKHYNLTVTCKRPGNKTVLPVTIMAGLVFHSQKYNLRLRQAWC
HFPSNWKGAWKEVKEEIVNLPKERYRGTNDPKRIFFQRQWGDPETANLWFNCHGEFFYCK
MDWFLNYLNNLTVDADHNECKNTSGTKSGNKRAPGPCVQRTYVACHIRSVIIWLETISKK
TYAPPREGHLECTSTVTGMTVELNYIPKNRTNVTLSPQIESIWAAELDRYKLVEITPIGF
APTEVRRYTGGHERQKRVPFVXXXXXXXXXXXXXXXXXXXXXXVQSQHLLAGILQQQKNL
LAAVEAQQQMLKLTIWGVK
The TM-Finder tolerates both spaces and numbers and is case insensitive. Any invalid characters (not valid amino acid symbols or other characters such as "%", "^", or numbers) that could have been entered by mistake are ignored by the program.
The sequence can be typed in the "Sequence" form or "cut" and "pasted" from a amino acids sequence source.
TM-Finder Training Sets
The TM-Finder program was optimized using data from a handful of membrane proteins whose structures have been determined to high resolution. The full sequences of this "Training-Set" of proteins are given in the "training sets" link below.
You may wish to use one or more of these proteins to optimize your choice of parameterization of the N-terminal window, C-terminal window, minimum core length, minimum segment length or the minimum gap length (These parameters are used by the TM-Finder to predict the TM-segments; changing these parameters may lead to a "better" prediction of the TM-segments).
The Training Set shows the experimentally determined TM-segments and also allows you to see the TM-Finder's prediction of the TM-segments in a sequence.
Return to the training sets »»
Running the TM-Finder
To run the program simply click the "run TM-Finder" button below the "Sequence" form. This outputs a graph showing the calculated hydrophobicity and helicity values, along with the predicted TM-segments if there are any.
For detailed information on the Graph, you could click the zoom-in, zoom-out, scroll-left or scroll-right buttons surrounding the TM-Graph. Clicking any of these buttons would bring up a more or less detailed graph (depending on which button you clicked).
The TM-Finder also has a re-scale option through which you can get the graph of a specified range of the sequence that you entered. To rescale the graph type in a start position in the "Start" box and and end position in the "End" box, and click the "Re-Scale Graph" Button. This brings up the re-scaled graph.
Saving TM-Finder Data
For use in a publication etc., a TM-Finder graph (in the results page) could be saved by pointing your mouse at the image and clicking the right mouse button. The hydrophobicity and helicity data of the sequence can also be obtained by clicking the "Hydrophobicity, Helicity" button and "TM-Segments" button near the bottom of the TM-Finder results page. This data is in CSV (comma separated values) format and can be used in a Lotus or Excel worksheet.
Amino Acid Sequences
The amino acid sequence can be any sequence composed of the 20 general amino acids.
An example of a sequence in FASTA format is:
&qt;gi|532319|pir|TVFV2E|TVFV2E envelope protein
ELRLRYCAPAGFALLKCNDADYDGFKTNCSNVSVVHCTNLMNTTVTTGLLLNGSYSENRT
QIWQKHRTSNDSALILLNKHYNLTVTCKRPGNKTVLPVTIMAGLVFHSQKYNLRLRQAWC
HFPSNWKGAWKEVKEEIVNLPKERYRGTNDPKRIFFQRQWGDPETANLWFNCHGEFFYCK
MDWFLNYLNNLTVDADHNECKNTSGTKSGNKRAPGPCVQRTYVACHIRSVIIWLETISKK
TYAPPREGHLECTSTVTGMTVELNYIPKNRTNVTLSPQIESIWAAELDRYKLVEITPIGF
APTEVRRYTGGHERQKRVPFVXXXXXXXXXXXXXXXXXXXXXXVQSQHLLAGILQQQKNL
LAAVEAQQQMLKLTIWGVK
This protein does not have a TM segment
Another example of a sequence in FASTA format is:
>gi|118508|sp|P17201|DHAQ_ACEPO MEMBRANE-BOUND ALDEHYDE DEHYDROGENASE (PYRROLOQU
INOLINE-QUINONE) PRECURSOR (ALDH)
MGRLNRFRLGKDGRREQASLSRRGFLVTSLGAGVMFGFARPSSANQIFPLDRSLPGDGAFEPTIWCSIAP
DGEITVNIIRAEMGQHIGTALARIIADEMEADWSKVRINYVDTDPKWGLMVTGGSWSVWMTWDVFRQAGA
ATRTAMVEEGARLLGTTPDKCTVASSIVSAGGKQISFGDIVAKGHPSHAFTPEEMAKLPLKPASERRLIG
NAELKALDIPAKTNGTAIYGIDAKVEGMLYGRPKMPPTRYGSKVRSVDDTEAKKIKGYVRYLLIDDPSQV
VQGWVVVLAESYSAAIRATDALKVEWTPGETIHTSERDIQDRGRELINNKAGGVYIFNDDGVDQAFGSAH
TVMDQEYTCASVLHYQLEPTNALAFEKDGVYEIHAGNQWQSLILPTLAKSLQVPESKVILRSYLLGGGFG
RRLNGDYMIPAALASKALGGKPVKLILTRSDDMQFDSFRSPSVQRVRMAFDASDRITAMDYQAAAGWPTG
VMAEAFMEKGVDGKPYDQFAIAGGDHWYEVGAFRVRALRNDLAEKTFRPGWLRSVSPGWTSWGVECFLDE
VAHRQKKDPAQFRLELLTGQGRNKGQAPDSVGGALRQAAVVRRLMEKVNWGKTSLPKDTAMGLATTAGQE
RGMPTWDRCVAQVHVDRSTGVVTCQKLTILVDAGTVVDPDGAKAQTEGAALWGLSMVLFENTEIVNGMPV
DRNLNTYTPLRIADTPEMDIEFLPSTEKPMGLGEPGTTVVGPAIGNAIFNAVGVRLRHMPVRPADVLRGL
QNG
This protein has a TM segment
When you are testing the TM-Finder program you may "cut and paste" these sequences into the TM-Finder to see how it works.
Return to the TM-Finder introductory page »»
Experiencing problems with this web tool? Contact