Transmembrane Helices Prediction Using Conditional Random Fields


 Enquiries go to: Wei Liu

To predict the transmembrane (alpha) helices in proteins, you would have to store the proteins primary sequences into a plain text file, and upload it onto this server.

Input Format:

Please make sure your uploaded file is in plain text format (.txt). The "protein ID line" should begin with "#", you can add you comments in this line. Below this line should be the corresponding protein primary sequence. Every amino acid in the sequence has to be represented by 1-Letter. The following is an example (two sequences):

#Sequence 0
MENLNMDLLYMAAAVMMGLAAIGAAIGIGILGGKFLEGAARQPDLIPLLRTQFFIVMGLVDAIPMIAVGLGLYVMFAVA
#Sequence 1
VQLAHHFSEPEITLIIFGVMAGVIGTILLISYGIRRLIKK

Or you can have a look at the sample primary sequences file, which is downloaded from the TMH Benchmark server.

Output Format:

The server uses "1" to denote Helix Segments, and "0" to denote Non-helix Segments. The following is an example (two sequences):

>0
0000000000111111111111111111111000000000000000000000111111111111111111111000000
>1
0000000000011111111111111111111111000000

To start prediction, upload your primary sequences (plain text file) onto the server:

Enter file name or click Browse...


© 2002-2010, the University of Sydney, NSW 2006 Australia