D03 Odsek za kemijsko informatiko

PredαTM and PredβTM: Transmembrane Region Predictors

PredαTM and PredβTM are two independent algorithms, which predict the transmembrane regions of integral membrane proteins. The underlying model is a SVM classifier trained on sequence data of transmembrane proteins with known structures. Given a protein sequence as input, the algorithms predict the probable transmembrane regions based on amino acid adjacency frequency and position specific preference of amino acids in the transmembrane regions.

For more information contact:
Please paste your protein sequence below or upload the input file. Choose whether to execute PredαTM or PredβTM algorithm. Your e-mail is necessary to be able to send the results.

*Number of sequences is limited to 10 per run. Extra sequences will be ignored automatically.
*Please make sure to enter the protein sequences in the correct input format.

The allowed input format is plain text or FASTA format.
A correct protein sequence should contain only the one letter abbreviations of the 20 natural amino acids. Missing residues can be represented by X. No other characters should be present in the input sequence.
Each protein sequence should follow a sequence identifier marked by > in a separate line.
Multiline field to enter protein sequence(s)

The input file should be a .txt file containing protein sequences in the correct format (see example).

Choose a script to execute:
PredαTM predicts the probable alpha helical transmembrane regions present in a given protein sequence of an integral membrane protein.

PredβTM predicts the probable beta strand transmembrane regions present in a given protein sequence of an integral membrane protein.
Choose an option to obtain the results:
e-mail (The results will be sent to this e-mail)
ID-number (The results file will be available 24 hours on the server, the file name is "ID-number.txt)