Rebuilding Z-DNA

[shr]

I was wondering if there is a way to rebuild Z-DNA with rebuild module in the local installation of x3DNA. I know it works for B-DNA and A-DNA and I can make a Z-DNA fiber model as well but I couldn't find a direct way to extend an existing Z-DNA structure. Is there a workaround which I can try? Thank you in advance.

[xiangjun]

Hi,

Thanks for using 3DNA and for posting your questions on the Forum. 3DNA rebuild should be able to build Z-DNA structures given a set of parameters. Please be specific with what you are trying to achieve, and we can start from there.

Best regards,

Xiang-Jun

[shr]

I have a PDB structure of Z-DNA-protein complex that I wanted to simulate. Before simulation, I wanted to make the DNA longer to avoid the issue of the protein interacting with any of the terminal bases. I can use the following commands to achieve this in case of B-DNA:

find_pair 1MNN.pdb 1mnn.bps
x3dna_utils cp_std bdna
rebuild -atomic bp_step.par extended_dna.pdb

But x3dna_utils does not take zdna as an argument so I was wondering how I can do it?

[xiangjun]

Thanks for your quick follow-up.

As noted in the x3dna_utils cp_std -h help message, the utility covers the most common use cases:

Select the standard data files to be used with "analyze" and "rebuild".
Available sets include BDNA, ADNA, NDB96 and RNA, which have exactly
the same base geometry and orientation (in the standard base reference
frame) but different backbone conformations.

Z-DNA is different from the standard right-handed DNA/RNA double helix in that it has not only a left-handed twist but also a base flip, and it has a di-nucleotide (most commonly CpG) as a the building block. So the x3dna_utils cp_std does not cover Z-DNA. However, you can run analyze, modify the output parameters (and extend as needed), and then rebuild a Z-DNA structure according to the modified parameters.

What specific Z-DNA structure you’d like to extend? If you do not want share details, please use a sample Z-DNA structure that helps illustrate your point. Reproducibility is important.

Best regards,

Xiang-Jun

[shr]

Thank you for your quick responses!
My input structure is the PDB structure 1QBJ. It is a 6nt DNA. I have attached a picture. Without using the x3dna_utils, even without extending, the backbone cannot  be modelled due to the distance of O3' and P being over 4.5 and throws the following error. However, I also tested with the fiber model of Z-DNA to understand how it works with Z-DNA. I found that besides the backbone not being modelled which is because of the PDB dataset not being since I'm not using x3dna_utils, rebuild could model the bases as separate molecules. So, this is possibly the issue with my Z-DNA crystal structure model.

Another observation was, previously, for the crystal structure of BDNA complexed with the protein, I tried to manually extend the DNA in pymol but due to the terminal base positions, I could not properly build the double helix. But after analyze and rebuild, I could. So I assume it is due to standardization of the structure after rebuild. However, there is a problem with this method for Z-DNA. Even if I manage to standardize the structure, pymol does not offer building Z-DNA. So, I was wondering if I could add the extended DNA before rebuild. As you suggested, I modified the bp_step.par file since that is the file used as input for rebuild. Is that the correct way? I apologize if I am asking too many questions.   

[xiangjun]

Thanks for your follow-up questions and the details you provided. It is always helpful to to be specific when discussing research topics.

Yes, "rebuild -atomic" would have issues with backbone connectivity, since in Z-DNA, nucleotide G is in syn conformation instead of anti (for C). The building block must be adjusted accordingly. I'll look into this further to see what we can get.

Another approach is to take the whole Z-DNA structure as a unit, and perform some transformations to extend it. See the PyMOL thread a few years ago on "create a 26 bp RNA from a 13 bp system" (https://www.mail-archive.com/pymol-users@lists.sourceforge.net/msg16190.html). The idea is applicable to Z-DNA as well. Note that the features are now available in the free DSSR Academic license (previously in DSSR Pro Academic only). Check if that method makes sense to you.

Best regards,

Xiang-Jun

[xiangjun]

Does the attached PDB file (with base schematic image) fulfill your needs? The backbone connection between the two segments are a bit longer than normal O--P covalent bond distance, which you can regulated with energy minimizations (e.g., using Phenix, as shown in "Web 3DNA 2.0 for the analysis, visualization, and modeling of 3D nucleic acid structures" (https://doi.org/10.1093/nar/gkz394).

[shr]

Thank you so much! But I am unable to properly download the pdb file. I will try to follow the method you mentioned about in your previous reply.

[xiangjun]

But I am unable to properly download the pdb file.

What do you mean? Just click on the link and it should download automatically. I've never heard of any issues with downloading files as long as you have an active internet connection. Please clarify your issue so I can assist you better.

[shr]

Sorry! I could download it now. It was probably an issue with my browser.

[Di_Liu]

Hi Xiang-Jun, following up on the Z-DNA rebuilding, is there a way to create a Z-DNA circle? I think the difficulty lies in how to rebuild to create a backbone of Z-DNA using the helical parameters. Thanks!

[xiangjun]

Hi,

Thanks for chiming in on the discussion. Currently, DSSR can build DNA circles with right-handed helices, but not Z-DNA forms. The backbone of Z-DNA is dramatically different from that of B-DNA or A-DNA, and needs to be modeled differently. I'll look into how we can incorporate Z-DNA backbones into DSSR modeling functionalities, given enough interest from the community, and with a proper collaborator to work on it. See the DSSR-Jmol and DSSR-PyMOL integration for two concrete examples of what I have in mind for such collaborations.

As for the current thread, I'm hoping @shr could respond to my question on March 15, 2025:

Does the attached PDB file (with base schematic image) fulfill your needs?

See my recent post On registration and posting which includes a copy of "Registration Agreement for the Forum" at the bottom.

Best regards,

Xiang-Jun

[shr]

Apologies for it took a while to reply. Thank you for your help with the structure! I used phenix to minimise like you suggested and I can use it for my analysis now. I would also look forward to Z-DNA backbones being included in DSSR modeling functionalities.

[xiangjun]

Apologies for it took a while to reply. Thank you for your help with the structure! I used phenix to minimise like you suggested and I can use it for my analysis now.

Thanks for the confirmation that the DSSR modeled Z-DNA structure works for your case. It is through interactions with real-world users like you that makes DSSR relevant and useful.

Even though virtual, a forum thread is just like a conversation. I strive to respond to users' questions timely and concretely. Users who initiate a thread are expected to follow up with their progress and share their findings, or the lack thereof. It’s a two-way street, and I appreciate your engagement and contributions to the forum.

As a follow up, I am planing to write a blogpost to summarize the discussion and provide step-by-step details on how the extended Z-DNA structure was built using DSSR. This may take a couple of days, and I will provide a link here once it is ready.

I would also look forward to Z-DNA backbones being included in DSSR modeling functionalities.

I have split a new thread titled "Rebuilding circular Z-DNA". Please share your thoughts and suggestions on this topic over there.

Best regards.

Xiang-Jun

[shr]

I apologize once again. I look forward to the blogpost and will check out the thread "Rebuilding circular Z-DNA". I would be happy to contribute in any way I can.

[xiangjun]

I apologize once again.

I appreciate your attitude and the effort you put into bringing up this interesting topic on the Forum. This case also reminds me that I should be a bit more proactive in engaging with users like you. I always respond directly to users' questions, no matter how minor. Whenever necessary, I ask for clarification to ensure I understand and address their issues correctly. However, if there’s no response, I normally shy away from pushing users further—even though I may have done extra work to make sure I’m on the right track.

I look forward to the blogpost and will check out the thread "Rebuilding circular Z-DNA".

Done. Please see the blogpost Building extended Z-DNA structures with backbones using DSSR. Following the steps in the blogpost, users should be able to build extended Z-DNA structures exactly as described.

I would be happy to contribute in any way I can.

I noticed your follow-up post and the one from @Di_Liu on the new thread Rebuilding circular Z-DNA under the section 'RNA structures (DSSR).' This kind of participation is exactly what I have in mind for the forum. I will get back to you over there soon.

Working together, we should be able to come up with a solution that not only solves the problem at hand but also benefits the community. DSSR would become even more powerful!

Best regards,

Xiang-Jun