Docking/Fitting a Domain (Atomic Resolution or Low Resolution) into a Cryo-EM Map using MOTIF-EM

Here we show how to use MOTIF-EM http://ai.stanford.edu/~mitul/motifEM/ to do docking/fitting of atomic resolution structures/domains (or even low resolution cryoEM structures) into low resolution structures from Cryo-EM or electron density maps.

Advantages of using MOTIF-EM to dock/fit compared to other software (as of 2011)

1. Partial Matching. That is, it is ok if only part of the domain, to be docked, fits in the map. This is MOTIF-EM’s biggest advantage over other software. It relieves the user of knowing domain boundaries.
2. It can be much faster that other docking/fitting software, because it can be run on a computing cluster comprising of many nodes.
3. You don’t have to specify an initial location or orientation. That is, it does global search.

Scope and Limitation (DO HEED THIS)

Same as for MOTIF-EM: http://ai.stanford.edu/~mitul/motifEM/compiling.html.

Preperation

At this point it is assumed that you have installed MOTIF-EM as per: http://ai.stanford.edu/~mitul/motifEM/compiling.html

Additionally, you will need to install EMAN [LBC99], as we will use the tool “pdb2mrc” from it.

Generate the input file “inp.txt” by running this command:

set_up_domain_fit.pl <map.xplor> <domain.pdb> <map_resolution> <work directory> <MOTIF-EM source directory>

- <map.xplor> is the input cryoEM map in XPLOR format

- <domain.pdb> is the input high resolution PDB file of a domain you want to fit in the map

- <map_resolution> is the resolution of the cryoEM map in Å

- <work directory> is the directory you want to work in

- <MOTIF-EM source directory> is the MOTIF-EM installation with full path

If the domain you want to fit is another cryo-EM map, then instead use this command:

set_up_map_fit.pl <map.xplor> <domain.xplor> <work directory> <MOTIF-EM source directory>

- <map.xplor> is the input cryoEM map in XPLOR format

- <domain.xplor> is the input low resolution domain you want to fit in the map

- <work directory> is the directory you want to work in

- <MOTIF-EM source directory> is the MOTIF-EM installation with full path

Running MOTIF-EM

Now, in the current working directory where “inp.txt” is located, run MOTIF-EM (the Step -4- of the MOTIF-EM running steps at: http://ai.stanford.edu/~mitul/motifEM/run_commands.html) 

Analyzing the results

Now many fits are generated. To see the first N fits, run:

generate_N_fits.pl <domain.pdb> N

- <domain.pdb> is the input high resolution PDB file of a domain

The fits are generated as “<domain.pdb>_fit_1.pdb”,  “<domain.pdb>_fit_2.pdb”, and so on.

Now, only a part of domain might have fitted onto the map. So <domain.pdb>_fit_i.pdb.part.pdb is that part of <domain.pdb>_fit_i.pdb. This shows the partial matching/fitting strength of MOTIF-EM.

However, to generate only the i-th fit, run:

generate_fit.pl i <domain.pdb> <domain.fitted.pdb>

- <domain.pdb> is the input high resolution PDB file of a domain

- <domain.fitted.pdb> will be domain.pdb fitted onto the input map as per the i-th solution.

If the domain you are fitting is another cryo-EM map, then instead use this command:

To see the first N fits, run:

generate_N_map_fits.pl <domain.xplor> N

- <domain.xplor> is the input low resolution domain you want to fit

The fits are generated as “domain.xplor_fit_1.xplor”,  “domain.xplor_fit_2.xplor”, and so on.

Now, only a part of domain might have fitted onto the map. So <domain.xplor>_fit_i.xplor.part.xplor is that part of <domain.pdb>_fit_i.xplor. This shows the partial matching/fitting strength of MOTIF-EM.

However, to generate only the i-th fit, run:

generate_map_fit.pl i <domain.xplor> <domain.fitted.xplor>

- <domain.xplor> is the input low resolution domain you want to fit

- <domain.fitted.xplor> will be domain.xplor fitted onto the input map as per the i-th solution.

References

[SLC10] Saha, M., Chiu, W. & Levitt, M. MOTIF-EM: an automated computational tool for identifying conserved regions in CryoEM structures. M. Bioinformatics 26 (12), 301-309 (2004).

http://cs.stanford.edu/~mitul/motifEM/

[LBC99] Ludtke, S.J., Baldwin, P.R., and Chiu, W. (1999). EMAN: semi-automated software for high-resolution single-particle reconstructions. J Struct Biol 128, 82-97.