MCMBB: Markov Chain Model for Beta Barrels

MCMBB is a fast algorithm, which discriminates beta-barrel outer membrane proteins from globular proteins and from alpha-helical membrane proteins. The algorithm is based on a 1st order Markov Chain model, which captures the alternating pattern of hydrophilic-hydrophobic residues occurring in the membrane-spanning beta-strands of beta-barrel outer membrane proteins. The model achieves high accuracy in discriminating outer membrane proteins, since it can discriminate beta-barrel outer membrane with a correct classification rate of  90.08% and the globular proteins with a correct classification rate of 92.67%. When submitting alpha-helical membrane proteins, the method shows an accuracy of 100%.

A score greater than zero, indicates that the protein is more likely to be a beta-barrel outer membrane protein, whereas a result lower than zero, indicates that the protein is probable not a beta-barrel.

You may enter up to 1000 sequences in Fasta format.

Please cite:
Bagos PG, Liakopoulos TD, Hamodrakas SJ.
Finding beta-barrel outer membrane proteins with a markov chain model.
WSEAS Transactions on Biology and Biomedicine, 2004, 2(1) 186-189.[PDF]