Computer analysis of nucleic acid sequences.

The increasing body of nucleic acid sequence data has created interest among many scientists in computational aspects of data storage and data processing. In fact GenBank and other data bases hade been created in order to store the data in a useful format, both for archival and analysis purposes. (See Burks et a/.' for a review of GenBank activities.) The value of simply have easy a m e s to alJ ribosomal RiVAs, for example, is not to be underestimated. The purpose of this article is to examine some of the array of tools that have been created in order to look at sequences in a rigorous, systematic way, utilizing the power of modern computes. These analyses k 3 a n in the early 1970s and are now becoming more j iwd on specific problems such as conscnsus patterns in regulatory xquencts or RNA folding. The sections below will outline soerne methods of computer anal)sh of sequence data The intent is to describe the analyses and to give emphasis to the possible utility of the analysis, not to present detailed mathematics or computer science of the techniques. Reference is made to papen w h m more technical dctaib of equations, pointers, and data svuctum arc to k found. A great many algorithms have been proposed in recent years. See the issues of Nucleic Aci& Reear&-' and the Bulletin of Marhematical Biology' devoted to computer methods for a good sample ofthis literature. Frequently ideas such as open reading fnme analysis or dot matrices arc rediscowred and reimplemented over and over. For this reason I make no attempt to survey the literature. Instead 1 uy to'descrik some useful and interesting metbods of sequence and+ that utilize the power of comp u t e s Sewd different questions might be asked about a sequence. One concerns unexpected relationships With other sequences: these discoveries are sometimes made by screening a'data base with the sequence; Wilbur