It might work using "catdcd" (http://www.ks.uiuc.edu/Development/MDTools/catdcd/). You'd need a textfile which just contains the atom numbers (zero-based) for your C-alphas. This file could be created in VMD using a TCL code such as: (untested)

set a [atomselect top "name CA"]

set atoms [$a get index]

set out [open "indexfile.txt" w]

puts $out $atoms

close $out

In order to be able to work with the resulting file, you might need a PDB file of only the CAs too:

$a writepdb "short.pdb"

Afterwards use catdcd e.g.:

catdcd -o short.dcd -otype dcd -i indexfile.txt -dcd large.dcd

If you know the structure of the binary DCD file, you can read it, and put DCD coordinates into a nested list with a Python script---see the file attached.

You can make a list of atom identifiers of Calpha atoms from your reference PDB file, filter the list of coordinates, and write a multi-PDB file.

It might work using "catdcd" (http://www.ks.uiuc.edu/Development/MDTools/catdcd/). You'd need a textfile which just contains the atom numbers (zero-based) for your C-alphas. This file could be created in VMD using a TCL code such as: (untested)

set a [atomselect top "name CA"]

set atoms [$a get index]

set out [open "indexfile.txt" w]

puts $out $atoms

close $out

In order to be able to work with the resulting file, you might need a PDB file of only the CAs too:

$a writepdb "short.pdb"

Afterwards use catdcd e.g.:

catdcd -o short.dcd -otype dcd -i indexfile.txt -dcd large.dcd

Norman's Answer has the best most efficient approach. Catdcd is a great tool. You should get yourself familiar with.

Dear Norman,

thank you for your detailed answer and introducing catdcd, it solved my problem. I just made some changes in command:

./catdcd -o instead of -ofile

Yes, you are right, I changed the answer correspondingly

Guess I confused the tags with antechamber ^^