Folding algorithm accuracy

The accuracy of the folding algorithm was tested on a database of structures that contains four small subunit ribosomal RNAs (rRNA), four Group I introns, three Group II introns, and 41 transfer RNAs (tRNA). The rRNAs and Group II introns were folded in domains of less than 500 nucleotides (5). Table III summarizes the percent of correctly predicted base pairs for this version of the algorithm and the last... 

The inherent variability of predictive success rate depending on the protein fold class brings important observations (1) When reporting accuracies the selection of the test set proteins should be balanced In order to include a representative number of each of the protein fold classes. (2) the prior knowledge of the protein fold class (Chou and Zhang, 1995) can be a valuable aid for the predictions and with that one can use the different algorithms in combination to predict a specific structural element of the chain. 

Over the years, there has been a sea of change in the attitude of chemists. Earlier, any theoretical method and the numbers computed from it had to be justified by comparing with experimental results. This has drastically changed because of two reasons First, the sophistication in theory, algorithms, and computer codes is now so good that these frequently deliver computed numbers much beyond present-day experimental accuracy. Second, there are many situations in which it is extremely difficult to perform an experiment, e.g., to study a very short-lived molecule or study a phenomenon like the folding of a protein in a biological environment. Theoretical and computational chemistry could be the only route to take in such cases. 

---