An algorithm for identifying regions of a DNA sequence that satisfy a content requirement

We present a dynamic progamming algorithm for identifying regions of a DNA sequence that meet a user-specified compositional requirement. Applications of the algorithm include finding C + G-rich regions, locating TA + CG-deficient regions, identifying CpG islands, and finding regions rich in periodical three-base patterns. The algorithm has an advantage over the simple window method in that the algorithm shows the exact location of each identified region. The algorithm has been implemented as a pol table C program called LCP (Local Content Program). LCP is extremely efficient in computer time and memory; it instantly locates all regions of a long DNA sequence meeting a given requirement. The LCP program was used to analyze the rabbit alpha-like globin gene cluster sequence.