Images and screenshots
Screenshots and other example images:
automatic multiple fitting using MultiFit
- Keren Lasker
(hint: click '+' to add more items)
=Input: emd_5137.map, 3LOS_A.pdb; Output: mmcpn_closed_state_model_multifit.pdb
=An automated protocol:
The density was split into two rings along the Z axis using simple map manipulations commands in the em module of IMP.
An ensemble of symmetric models for the top ring were generated using MultiFit. These models maximize fit to the map ,geometric complementarity between interacting subunits and maintain C8 symmetry.
The top ranked model, was refined using the SymmRef method.
The refined model was the fitted to the bottom ring, and the combined model of the two rings was deposited.
=Running times (on a single CPU):
Step 2 run for 4:41 minutes
Step 3 run for less then a minute
=The commands used to run multifit:
Generate a parameters file: $IMP/tools/imppy.sh python $IMP/modules/symm_multifit/bin/build_symm_multifit_params.py -- 8 3LOS_A.pdb emd_5137_top.mrc 4.3 1.33 0.468 -128 -128 -128
Generate a surface file for the intput subunit: $IMP/tools/imppy.sh perl $IMP/modules/symm_multifit/bin/runMSPoints.pl 3IYF_A.pdb
Run the modeling $IMP/modules/symm_multifit/bin/symm_multifit multifit.param.top
The output file (multifit.output) holds 10 top solutions (number of solutions is a parameter), each containing transformations to build the ring and the cross-correlation score. The records are sorted by the cross-correlation score.
alternatively, one can use the webserver: http://modbase.compbio.ucsf.edu/multifit/
Submitted PDB Models: