A new decoding algorithm for hidden Markov models improves the prediction of the topology of all-beta membrane proteins
2005

New Algorithm Improves Prediction of Membrane Protein Structure

Sample size: 20 publication Evidence: high

Author Information

Author(s): Fariselli Piero, Martelli Pier Luigi, Casadio Rita

Primary Institution: Department of Biology, University of Bologna

Hypothesis

Can a new decoding algorithm improve the prediction of the topology of beta-barrel membrane proteins?

Conclusion

The new PV decoding algorithm outperforms existing methods in predicting the topology of beta-barrel membrane proteins.

Supporting Evidence

  • The PV decoding achieved 80% accuracy in cross-validation.
  • The PV algorithm is more effective when multiple paths have similar probabilities.

Takeaway

Scientists created a new way to guess the shape of certain proteins that live in cell membranes, and it works better than older methods.

Methodology

The study tested a new decoding algorithm called posterior-Viterbi (PV) on a set of beta-barrel membrane proteins using a hidden Markov model.

Limitations

The algorithm may not perform better than others when a single best path dominates.

Digital Object Identifier (DOI)

10.1186/1471-2105-6-S4-S12

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication