Albany 2001

category image Biomolecular
SUNY at Albany
June 19-23, 2001

Identification of CAP Binding Sites Using Hidden Markov Models

Hidden Markov Models (HMMs) provide a powerful machine learning approach that has been used to identify subtle trends in DNA (1, 3) and proteins (2). HMMs trained on e. coli catabolite activator protein (CAP) binding sites distinguishes by means of Viterbi log odds scores between known CAP binding sites not included in the training set and non-binding sites. We hope to extend this model to the identification of such binding sites based directly on the bendability of the DNA binding sites, as this is a possible means of DNA recognition by proteins.

References and Footnotes
  1. Baldi, P., Brunak, S., Chauvin, Y., and Krogh, A. (1996) "Naturally Occurring Nucleosome Positioning Signals in Human Exons and Introns." J. Mol. Biol. 263, 503-510.
  2. Bateman, A., Birney, E., Durbin, R., Eddy, E. R.,Howe, K. L. and Sonnhammer, E. L. L. (2000) "The Pfam protein families database." Nuc. Acids Res. 28, 263-266.
  3. Pedersen, A. G., Baldi, P., Chauvin, Y., and Brunak, S. (1998) "DNA Structure in Human RNA Polymerase II Promoters." J. Mol. Biol. 281, 663-673.

Kelly M. Thayer (1) and David L. Beveridge (2)

Department of Molecular Biology and Biochemistry & Molecular Biophysics Program(1), Department of Chemistry & Molecular Biophysics Program(2), Hall-Atwater Laboratory, Wesleyan University, Middletown, CT 06457
Phone: (860)-685-2575; Fax: (860)-685-2141; Email: dbeveridge@mail.wesleyan.edu