FOLD RECOGNITION USING PREDICTED SECONDARY STRUCTURE SEQUENCES AND HIDDEN MARKOV-MODELS OF PROTEIN FOLDS

We present an analysis of the blind predictions submitted to the fold recognition category for the second meeting on the Critical Assessment of techniques for protein Structure Prediction. Our method achieves f old recognition from predicted secondary structure sequences using hid den Markov models (HMMs) of protein folds, HMMs are trained only with experimentally derived secondary structure sequences of proteins havin g similar fold, therefore protein structures are described by the mode ls at a remarkably simplified level. We submitted predictions for five target sequences, of which four were later found to be suitable for t hreading. Our approach correctly predicted the fold for three of them, For a fourth sequence the fold could have been correctly predicted if a better model for its structure was available, We conclude that we h ave additional evidence that secondary structure information represent s an important factor for achieving fold recognition. (C) 1998 Wiley-L iss, Inc.