Structural and combinatorial constraints on base pairing in large nucleotide sequences

In this paper we discuss the constraints and combinatorial problems of folding long RNA and single stranded DNA molecules into base paired structures. A computer code FOLD-A was designed to perform base pairing foldings of very long sequence chains and search for low energy configurations. The logic of the FOLD-A algorithm is described in some detail. The applications of FOLD-A to the A-protein gene of MS2 and the whole genome of the phi X 174 phage with over 5300 bases are discussed in the accompanying paper.