A microcomputer program for comparison and alignment of DNA sequence gel readings.

During the course of determining the sequence of a large DNA fragment, it is necessary to cross-check numerous gel readings from different DNA fragments, in order to track and eliminate mistakes. An algorithm is presented that takes advantage of the high degree of homology between such sequences to construct an alignment of the matching regions. It does not require knowledge of a starting homology zone, neither large memory areas, even for sequences of several kilobases, and it can overcome large gaps or mismatch zones that correspond for instance to misinterpretation of compressions on sequence gels. This algorithm has been implemented in 6502 assembly language on an Apple II computer as an extension to the PEGASE sequence handling system.