How to control the direction in which the two DNA chains are concatenated

[Shuming_Liu]

Hi everyone,

These days I am trying to build a model by concatenating two DNA chains together. I followed this tutorial: http://forum.x3dna.org/users-contributions/build-dna-bulges-and-extend-dna-duplex-at-both-terminals-via-3dna/

I have used find_pair and frame_mol to align the ends of two DNA chains. However, I have met the problem that I cannot control the direction in which they are concatenated. That is to say, I imagine after concatenating them together by overlapping the ending base pairs, both chains will extend in different directions; but the result is that the two chains are extending in the same direction and largely overlaps. I don't know how to control the direction in which the endings of two chains are aligned. I have tried to change chain ID as well as change residue ID, but they also do not work. So I am wondering if someone has experience in solving such problems, thanks!

[xiangjun]

Please be specific with detailed steps you took and problems you experienced.

Thanks,

Xiang-Jun

[Shuming_Liu]

Hi Xiang-jun

I am now trying to build a poly-nucleosome model, so I have to concatenate mononucleosomes. My script is like this:

 find_pair fiber-177-2.pdb stdout # fiber-177-2.pdb is the dinucleosome structure
 frame_mol -353 ref_frames.dat fiber-177-2.pdb fiber-177-2_aligned.pdb
 find_pair block-178.pdb block-178.bps # block-178.pdb is a mononucleosome
 frame_mol -177 ref_frames.dat block-178.pdb block-178_aligned.pdb

So I am trying to concatenate block-178.pdb to fiber-177-2.pdb, and I have enclosed a picture to show the result. You can see that in the red box region of the picture, two DNA chains are overlaping at the ending, but extending to the same direction. But what I want is to link these two DNA chains together (by overlaping only the ending base pair) and they should extend in opposite directions. I just want to know how to fix this, thanks!

[xiangjun]

Would it be feasible that you attach all the PDB files? Without those data files, it is impossible to reproduce the reported image and it is hard (for me at least) to pinpoint exactly where the issue is.

Xiang-Jun

[Shuming_Liu]

Hi Xiang-jun

Thank you for your reply! Since as for now it might be inconvenient for us to share the whole model, I use a toy model as an example, and I reproduce this problem on the toy model. I want to concatenate part-1.pdb and part-2.pdb by overlapping residue 13 and 10 (a base pair) of part-1.pdb and residue 20 and 328 (another base pair) of part-2.pdb. I use the script run.sh to align two molecules into part-1_aligned.pdb and part-2_aligned.pdb. But the result, same as before, the two chains are extending in the same direction, which is not what I want. So I am wondering how to concatenate them in the true direction so that they can extend in the opposite direction. Thank you very much!

[xiangjun]

Hi,

Thanks for providing a detailed example. It is completely fine, even preferable, to use a simplified data set just to illustrate the issue at hand.

With the data files and steps you provided, I can now understand clearly where the issue is. The 3DNA programs are working as expected, but one step is missing. Please try the following:

Code: [Select]

# Create a text file ('rot_yz180', attached), with the following content:
by rotation y 180
by rotation z 180

# transform file 'part-2_aligned.pdb' (attached), as below
rotate_mol -r=rot_yz180 part-2_aligned.pdb part-2_aligned-rotated.pdb
Now use part-2_aligned-rotated.pdb instead of part-2_aligned.pdb in your concatenated file. Have a try and report back how it goes.
 
Xiang-Jun

[Shuming_Liu]

Hi Xiang-jun,

Thank you very much! This method does solve the problem, not only on the toy model, but also on the model we are trying to build. Thank you very much! This could be a great supplement to the method of concatenating DNA chains with 3DNA.