Pattern-Recognition Importance Sampling Minimization PRISM

Pattern-Recognition Importance Sampling Minimization (PRISM) 

The polypeptide chain is built up from the N terminus, fitting the most probable tripeptide conformations together, one tripeptide at a time, allowing for proper overlap of the tripeptides. As the build-up proceeds, the probabilities of the growing chain (conformation) are calculated, and only the 1000 most probable are retained. Thus, when the C terminus is reached, there are 1000 different predictions of the backbone structure of the protein, sorted in order of decreasing probability. 

---