The .mat files in this directory show the similarity scores
associated with the various -s abbrev scoring options for the
fasta36 programs. Thus, -s BP62 uses the BLOSUM62 matrix, which
is provided in blosum62.mat. There are four families of matrices provided in this directory:
-
The
pam120.matandpam250.matmatrices derived from the original Dayhoff (1978) PAM data (Dayhoff et al. (1978), Atlas of Protein Sequence and Structure, Vol. 5 suppl 3 pp. 345-352; National Biomedical Research Foundation, Silver Spring, MD -
The BLOSUM series of matrices (
blosum50.mat,blosum62.mat, etc., Henikoff and Henikoff (1992), PNAS 89:10915-10919) -
The MD series of matrices (
md_10.mat,md_20.mat,md_40.mat) described by Jones, Taylor, and Thornton (1992) Comp. Appl. Biosci 8:275-282. -
A series of matrices developed using the VTML model (Mueller and Vingron (2002) Mol. Biol. Evol. 19:8-13):
VTML_10.mat,VTML_20.mat, etc. -
In addition, the Optima5 matrix (Kahn and Goldstein (2002) Proteins 48:367-376) is provided (
-s OPT5).
The mapping of -s abbrev to data/*.mat file is found in
src/upam.h file in the definition of the struct std_pam_str std_pams[]
file:
| Abbrev. | .mat file |
|---|---|
VT10 |
vtml_10.mat |
P10 |
md_10.mat |
M10 |
md_10.mat |
MD10 |
md_10.mat |
VT20 |
vtml_20.mat |
P20 |
md_20.mat |
M20 |
md_20.mat |
MD20 |
md_20.mat |
VT40 |
vtml_40.mat |
P40 |
md_40.mat |
MD40 |
md_40.mat |
VT80 |
vtml_80.mat |
BL80 |
blosum80.mat |
VT120 |
vtml_120.mat |
P120 |
md_10.mat |
BL62 |
blosum62.mat with -8/-1 gap penalties |
BP62 |
blosum62.mat with -11/-1 gap penalties |
VT160 |
vtml_160.mat |
BL50 |
blosum50.mat |
OPT5 |
optima5.mat |
PAM250 |
pam250.mat |
In addition, several other scoring matrices are provided, including:
.mat |
|
|---|---|
idn_aa.mat |
identity matrix for protein searches |
dna.mat |
+5/-4 matrix for DNA searches |
rna.mat |
+5/-4 matrix for RNA searches with +2 for G:A/TU:C |