Back to Filters page.
Uses the sequence in the pose to generate secondary structure predictions for each position. Secondary structure predictions are then compared to the desired secondary structure to determine a score. If use_probability is true, the score returned is a value between 0 and 1, where 0 is complete secondary structure agreement, and 1 is no agreement. The following equation is used to determine the score: sum(i=1;N;e^(-p[i]/T)), where N is the number of residues, p[i] is the probability of correct secondary structure at position i, and T is a temperature factor set to 0.6 by default. If use_probability is false, the filter returns the fraction of residues that match the desired secondary structure as a number between 0 and 1. If use_probability is true AND mismatch_probability is true, the score is the geometric average of the probability of picking the WRONG secondary structure type at all residue positions. Minimizing this number will maximize the geometric average of the probability of picking the CORRECT secondary structure type at all residue positions. This option should be the most correct method for comparing two sequences to determine their expected fragment quality based on their predicted secondary structure probabilities at each residue.
<SSPrediction name="(&string)" threshold="(&real)" use_probability="( false &bool)" mismatch_probability="( false &bool)" cmd="(&string)" blueprint="( '' &string)" use_svm="( true &bool )" />