3DCOMB result pdb chain to input structure mapping

[suyang-nju]

After running 3DCOMB, how do I find out to which input structure each segment/chain of the aligned structures correspond? Thanks!

[realbigws]

In DeepAlign package, you may find an example to run 3DCOMB by typing:

cd examples/3DCOMB_example ./3DCOMB -i input_list

Here, you may find that there are four structures in input_list, and the corresponding four chains in the resultant superimposed file input_list.pdb .

They are in one-to-one correspondence.

Hope this helps.

Best, -Sheng

On Wed, Nov 4, 2020 at 8:52 AM suyang-nju [email protected] wrote:

After running 3DCOMB, how do I find out to which input structure each segment/chain of the aligned structures correspond? Thanks!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/realbigws/DeepAlign/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACD6EWKLTNQYLMWX7TBB44TSOCQVTANCNFSM4TJNJIFA .

[suyang-nju]

Is the order of the structures (chains) that appear in the superimposed pdb file the same as they appear in the input_list? In other words, does the first structure in input_list appears first in the superimposed pdb file, the second structure in input_list appears second in the superimposed pdb file, etc? My input_list contains ~100 structures; the chain ids and segment ids in the superimposed pdb file do not appear to follow an obvious pattern. So I guess I have to iterate over the entire superimposed pdb to establish the mapping between input_list and the chain ids and segment ids in the superimposed pdb, and there's no quick way to calculate what the chain id and segment id would be in the superimposed pdb for the Nth structure in the input_list.

Thanks! Yang

[realbigws]

On Thu, Nov 5, 2020 at 2:41 AM suyang-nju [email protected] wrote:

Is the order of the structures (chains) that appear in the superimposed pdb file the same as they appear in the input_list? In other words, does the first structure in input_list appears first in the superimposed pdb file, the second structure in input_list appears second in the superimposed pdb file, etc? My input_list contains ~100 structures;

I strongly suggest that you input less than 25 structures. This is because of the fact that 3DCOMB is not that accurate to deal with more than ~30 input proteins for multiple structure alignment.

Best, -Sheng

the chain ids and segment ids in the superimposed pdb file do not appear to follow an obvious pattern. So I guess I have to iterate over the entire superimposed pdb to establish the mapping between input_list and the chain ids and segment ids in the superimposed pdb, and there's no quick way to calculate what the chain id and segment id would be in the superimposed pdb for the Nth structure in the input_list.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/realbigws/DeepAlign/issues/2#issuecomment-721905457, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACD6EWMXP7I6EQDZCPEGKN3SOGN5RANCNFSM4TJNJIFA .

[suyang-nju]

I see. Thanks for the suggestion.