next up previous contents
Next: Insertions and Deletions Up: Point Accepted Mutations and Previous: Estimating Mutation Matrices

Other Similarity Matrices

Darwin has several other matrices available to interested users. These can be loaded with the function Matrices.

> Matrices();

BLOSUM50, 60, 62, 70. Sequence similarity among protein sequences is typically measured by similarity matrices, such as the Dayhoff PAM matrices based on evolutionary rates, which give scores for all possible exchanges of one amino acid with another. Henikoff and Henikoff have derived substitution matrices using about 2000 blocks of aligned subsequences from 504 groups of related proteins. In FASTA and BLAST searches, these matrices have shown improvements in alignments over the other substitution matrices, using query sequences from each of the 504 groups [19].

PAM0, 50, 100, 150, 200, 250. Various ``enhanced'' Dayhoff matrices computed for different PAM values. See §[*].

Dayhoff1978. The original mutation matrix given in by Dayhoff et. al. [].

GENETIC10, 50, 100, 150, 200, 250. These matrices are derived directly from the genetic code. They assume that the only source of mutations are random mutations in any of the three bases [15].

Grantham. Grantham's structural dissimilarity matrix.

Machlaclan.

PIMA.

RDDH250. See the paper Amino acid subsitutions in structurally related proteins: a pattern recognition approach, [24].

UNITARY, UNITARY2.


next up previous contents
Next: Insertions and Deletions Up: Point Accepted Mutations and Previous: Estimating Mutation Matrices
Gaston Gonnet
1998-09-15