Euclidian Model Matching
EMMA stands for Euclidean Model Matching and is the algorithm used
by HySS to superimpose two putative solutions and to derive the
consensus model. The same algorithm is also available through the
external phenix.emma command-line interface. Enter phenix.emma
without arguments to obtain the help page:
usage: phenix.emma [options] reference_coordinates reference_coordinates other_coordinates
-h, --help show this help message and exit
External unit cell parameters
External space group symbol
--symmetry=FILENAME External file with symmetry information
--tolerance=FLOAT match tolerance
Use only if models are diffraction-index equivalent.
Example: phenix.emma model1.pdb model2.sdb
The command takes two coordinate files in various formats (.pdb, CNS
.sdb, SOLVE output, SHELX .ins) and compares the structures taking the
space group symmetry, the allowed origin shifts and the hand ambiguity
into account. The output is similar to the Match summary shown
above in the example HySS output.
The match tolerance defaults to 3 Angstrom. For structures obtained
with very low resolution data it may be necessary to specify a different
tolerance, e.g. --tolerance=5.
The --symmetry option works just like it does for phenix.hyss.
It can be used to extract symmetry information from external files such
as input files for other programs (CNS, SHELX, SOLVE, ...) or
reflection files. However, the --symmetry option is only required
if the information about the unit cell and the space group is missing
in both coordinate files given to phenix.emma.
phenix.emma conducts an exhaustive search and, in contrast to HySS,
displays all possible matches. The match with the largest number of
matching sites is shown first, the match with the smallest number of
matching sites is shown last (often just one site). Therefore you have
to look at the beginning of the output to see the best match. I.e. if
the output goes to the screen don't let yourself get distracted if you
see a large number of Singles near the end of the output. Scroll
back to see the best match.
Emma is also available via a web interface.