A computer method for construction of secondary structure from polynucleotide sequence. Possible structure of the...

A computer method to search the possible secondary structure of a long polynucleotide was developed. As a criterion for the stabilization of a secondary structure, free energy originating from base-pairing was employed, since the structure in solution would be at the free energy minimum. The method is summarized as follows: all possible helices are collected from a given nucleotide sequence under restrictions that the length of a helix is greater than N0 bases (e.g., four bases) and the free energy of the helix calculated according to free energies of two successive sequence-dependent basepairs is lower than E0 (e.g., -5 kcal/mol). The search of secondary structures of low free energy is performed by connecting one helix to another without allowing any base-pairing between loops. For connecting single-stranded regions, destabilizing free energy of 2--3 kcal/mol is added. The method was first applied to several tRNAs and the clover-leaf structure of tRNA was obtained as a free energy minimum. Then, possible secondary structures of the replication origin regions of the Escherichia coli and Salmonella typhimurium chromosomes were examined by the method, assuming that one of the strands in the origin region takes a specific secondary structure. The lowest-energy structure for the E. coli origin was found to be approximately identical to that for the S. typhimurium origin region.