Preliminary Pairwise Alignments


This a preliminary page with the raw, automatically generated alignments described in the paper below. There are future plans to present these alignments in an easier to use manner perhaps through a server. The actual results in the paper involve merging these many files into a single one. This will be done in the web presentation in a bit.

M Gerstein & M Levitt (1997). "Comprehensive Assessment of Automatic Structural Alignment against a Manual Standard, the Scop Classification of Proteins," Protein Science 7, 445-456.

Alignment Files

[ ] loop-sfam.00.gz
[ ] loop-sfam.04.gz
[ ] loop-sfam.40.gz
[ ] loop-sfam.44.gz
[ ] loop-sfam.50.gz
[ ] loop-sfam.70.gz

Files are compressed.
Number such as "00" or "04" gives the alignment method, using the following cryptic scheme. Interpret the number as a binary number. Then from left to right, the bits indicate: 

sec. str. dependent gap pen. ?
exp(cos theta) for CACBvec calc. ?
use CAs instead of CBs ?
use C=O instead of CA-CB ?
read in weights ?
verbose show input ?
simple dyn. prog.  ?
As described in the paper, not all alignment methods work on every scop pair, so one has to pick the best alignment for each pair.

Output format

ALIGN d1copd_ d1octc2
COMMON      >-------#########----#################--##################################----
d1copd_     >-------MEQRITLKD----YAMRFGQTKTAKDLGVYQSAINKAIHAGRKIFLTINADGSVYAEEVKPFPSNKKTTA
d1octc2     >DLEELEQFAKTFKQRRIKLGFTQGDVGLAMGKLYGND--FSQTTISRFEALNLSFKNMCKLKPLLEKWLNDAE----
_NENTRY   50  (number of equivalences)
_RMSBES    7.218 (best RMS)
_ASCORE   14.436 (best SAS score)
_MSCORX    140 (best matrix score)
_NBRKSX    2  (best number of breaks)
_SUM1 d1copd_ d1octc2  50   7.218  14.436
_SUM2 d1copd_ d1octc2 %_ bads   1   50   7.22    .00    .00   2    71    .00   7.22  12.03    .93

Format is two scop id's after ALIGN.  Actual alignment follows.  Then
RMS, Number of aligned residues, and structural alignment score.

[home]   Lab Home / Census Home