GCG: Production of Multiple Sequence Alignment

Abstract

Multiple sequence alignment is one of the most challenging lems in biocomputing. Sequence comparison, averaging, and so ticated editing are required to make the computer do wha researcher wants to do. However, frequently complaints are that an alignment is “bad” or “insufficient” if calculated auto cally. This is because of the fact that the computer does only pair alignments and comparisons, and compiles these results in a se step to produce the multiple sequence alignment. Further, the resea implies automatically which regions are to be “conserved” (i.e. to be interrupted by gaps or misaligned) and which sequence ments are less essential. Computationally, only match and mism scores are known at the beginning of an alignment, and feature conserved regions are established only during the program run.