GAMBIT MacClade TreeView

of various phylogenetics methods are summarized in useful reviews of phyloge-netics [74-77]. There are three common methods used in phylogenetic analysis, the distance matrix, maximum likelihood, and parsimony methods. The distance matrix method is faster and most commonly used in phylogenetic analysis. Parsimony uses the position-specific information in a multiple sequence alignment. It is sensitive but takes a longer time to run compared to the distance matrix method. Maximum likelihood uses a different model. It takes into account every sequence change, and it is slowest of the three methods. Table 7 gives information about the sources of information in the web for some of the phylogenetics programs.

B. Structure-Based Methods

Structure is evolutionarily conserved to a greater extent than sequence. Even if two sequences do not share obvious sequence similarity, they might share the same 3-D structure (fold). Thus, fold recognition methods have tremendous potential in characterizing unknown proteins. The following flowchart gives the consequences of evolution and the physicochemical properties of protein on the number of genes, sequences, and structures (adapted from Ref. 78).

>105 genes ^ 104 sequence families ^ ~103 folds

Although there are many genes, the number of domain sequence families is much smaller. The number of folds is almost certainly an order of magnitude less, as is the number of architectures. The architecture refers to the packing of sheets and helices in a structure regardless of sequential connectivity. At the end of the equation, the number of structural supersecondary motifs that constitutes the fold is very small. Biological complexity is achieved by using local variations together

^ ~102 structures ^ ~10 supersecondary motifs

Was this article helpful?

0 0

Post a comment