FORESST: fold recognition from secondary structure predictions of proteins

Motivation: A method for recognizing the three-dimensional fold from the pr otein amino acid sequence based on a combination of hidden Markov models (H MMs) and secondary structure prediction was recently developed for proteins in the Mainly-Alpha structural class. Here, this methodology is extended t o Mainly-Beta and Alpha-Beta class proteins. Compared to other fold recogni tion methods based on HMMs, this approach is novel in that only secondary s tructure information is used. Each HMM is trained from known secondary stru cture sequences of proteins having a similar fold. Secondary structure pred iction is performed for the amino acid sequence of a query protein. The pre dicted fold of a query protein is the fold described by the model fitting t he predicted sequence the best. Results: After model cross-validation, the success rare on 44 test proteins covering the three structural classes was found to be 59%. On seven fold p redictions performed prior to the publication of experimental structure, th e success rate was 71%. In conclusion, this approach manages to capture imp ortant information about the fold of a protein embedded in the length avid arrangement of the predicted helices, strands and coils along the polypepti de chain. When a more extensive library of HMMs representing the universe o f known structural families is available (work in progress), the program wi ll allow rapid screening of genomic databases and sequence annotation when fold similarity is not detectable from the amino acid sequence.