Recent Posts

11

Bug reports / Re: When analyzing multiple frames with --nmr, DSSR gets slower with each frame

« Last post by rkpoppleton on June 13, 2025, 09:58:08 am »

I see.  That makes sense where the slowdown is coming from.  For what its worth, in tools I've worked on in the past, when we need to do random-access trajectory reading, we first read through the whole trajectory once and make an index of the byte offsets for the start of each frame.  That way you can jump to any arbitrary frame without having to search through the file each time.

12

RNA structures (DSSR) / Re: Using --helical-axis with the --nmr option in DSSR

« Last post by rkpoppleton on June 13, 2025, 08:52:05 am »

Thank you!  The --more option was what I was missing.

13

RNA structures (DSSR) / Re: Using --helical-axis with the --nmr option in DSSR

« Last post by xiangjun on June 12, 2025, 02:57:39 pm »

Hi,

Thanks for your kind words about DSSR. Please try the --more option which will output "helical-axis", "point-one" and "point-one" as documented in the DSSR User Manual. You could parse the corresponding output for each model and start from there to calculate the bending angle.

Hope this helps.

Xiang-Jun

14

Bug reports / Re: When analyzing multiple frames with --nmr, DSSR gets slower with each frame

« Last post by xiangjun on June 12, 2025, 02:51:55 pm »

Hi,

Thanks for using DSSR and for posting your question on the Forum. I am aware of the issue you are describing. The initial design of the --nmr option allows for flexibility of user-selected frames/models (e.g., --nmr=3+5+6:9 as described in the User Manual). For each frame/model, DSSR re-reads the input file from the beginning, which causes the slowdown as you observed. There is no memory leak, as you can verify with valgrind or similar tools.

DSSR Pro version allows for sequential processing of all the frames in a single pass, which leads to faster performance (scale linearly with the number of frames).

Best regards,

Xiang-Jun

15

Bug reports / When analyzing multiple frames with --nmr, DSSR gets slower with each frame

« Last post by rkpoppleton on June 12, 2025, 08:48:31 am »

When I analyze MD trajectories with the --nmr option, the runtime does not scale linearly with the number of frames as one might expect.  For the same structure here are some benchmarks:

n_frames	runtime (hh:mm)
2500	1:00
5000	4:15
10000	>15:00

This suggests to me that there is either some sort of memory leak or maybe the file writing is seeking to the end of the file at the end of each frame?

Here is my DSSR command:

Code: [Select]

~/software/x3dna-dssr -i=RNA_traj.pdb --nmr --json -o=dssr.json
DSSR version: 2.4.6-2024nov15

16

RNA structures (DSSR) / Using --helical-axis with the --nmr option in DSSR

« Last post by rkpoppleton on June 12, 2025, 08:27:14 am »

Hi!  Thanks for the excellent tool, DSSR has solved so many problems in my analysis of RNA MD simulations!  I am currently trying to calculate the angle between two helices in an RNA structure.  For any single configuration, I can use the --helical-axis option to get the helical axes as a .pdb file, which works very well.  However this file seems to be overwritten for every frame if you run DSSR on multiple frames with the --nmr option.  I looked around in the --json output, but I can't seem to find a per-model output there. 

Are the helix axes output anywhere and I just missed them, or is there something else I need to do to get the dssr-helicalAxes.pdb file on a per-model basis?

17

RNA structures (DSSR) / Re: Rebuilding circular Z-DNA

« Last post by xiangjun on June 03, 2025, 12:42:43 am »

Hi Di,

Thanks for sharing the detailed steps you used to build the Z-DNA circle. What you called Y-shift and Z-shift are Slide and Rise, respectively, in the literature. Slide is an important parameter in determining DNA shapes, see "A Novel Roll-and-Slide Mechanism of DNA Folding in Chromatin: Implications for Nucleosome Positioning" and "The shape of the DNA minor groove directs binding by the DNA-bending protein Fis". It is also crucial for producing circular DNA structures, as you noticed.

Given below is the DSSR commands I used to generate the ZDNA-circle.pdb file I posted previously, plus further steps to improve the visualization of the 3D structure.

Code: Bash

1. # The starting point is your twist60-G84-scaled.pdb. Here only G on chain A is selected.
2. x3dna-dssr -i=twist60-G84-scaled.pdb --select-chain=A -o=chainA.pdb
3.  
4. # Extract a GpC step in Z-DNA conformation, and re-orient it in the reference frame of the first G (on Chain A)
5. x3dna-dssr fiber --z-dna --repeat=1 -o=fiber-GpC.pdb
6. x3dna-dssr -i=fiber-GpC.pdb --frame=A.1 -o=frame___Z.pdb
7.  
8. # Now mutate each G to a Z-DNA GpC step (frame___Z.pdb). The --mutate-type option is new in DSSR v2.5.3
9. #      "whole" to include backbone, and "raw-id" to keep the original identification of the atoms
10. x3dna-dssr mutate -i=chainA.pdb --entry="name=G to=Z" -o=ZDNA-circle.pdb --mutate-type=whole-raw-id
11.  
12. # The following steps would lead to better visualization of 3D structures
13. x3dna-dssr --order-residue -i=ZDNA-circle.pdb -o=temp_order.pdb --po-bond=3.6
14. x3dna-dssr --renumber-residue -i=temp_order.pdb -o=temp_renum.pdb
15. x3dna-dssr --connect-file -i=temp_renum.pdb -o=ZDNA-circle2.pdb --po-bond=3.6

The ZDNA-circle2.pdb and PyMOL-rendered image are attached. For completeness of this post, I have also attached twist60-G84-scaled.pdb from your previous post. With DSSR v2.5.3, users should be able to follow the above steps, and reproduce the results without any issues.

The overall strategy should be clear: in essence, the commands simply replace the 84 Gs with GpC dinucleotides steps in Z-DNA conformation. The method is generally applicable to other DNA/RNA modeling tasks, as demonstrated in my blogpost mentioned earlier "Mutate backbone of DNA and RNA structures". The integrative nature of DSSR is a key strength, and the automation it enables stands out when compared with alternative tools.

There are still areas that require refinement. I am more than willing to enhance the modeling capabilities in future releases of DSSR based on your feedback.

Best regards,

Xiang-Jun

18

RNA structures (DSSR) / Re: Rebuilding circular Z-DNA

« Last post by Di_Liu on June 02, 2025, 04:44:26 am »

Hi Xiang-Jun,

I've included my detailed procedure in this Evernote for your review: https://share.evernote.com/note/b6bd8283-bcad-58aa-0725-95dc61a9c6e0

I hope you can help explain why Y-shift is introduced when generating your B-DNA ring. Also, I think you would have a better idea than me of creating a Z-DNA ring with higher accuracy by taking into account the spatial relationship of the helical axis and the bp.

Thanks,

Di

19

RNA structures (DSSR) / Re: Rebuilding circular Z-DNA

« Last post by xiangjun on June 02, 2025, 12:50:18 am »

Hi Di,

Thanks for your confirmation. See my blog post Mutate backbone of DNA and RNA structures. The x3dna-dssr mutate sub command can now mutate a base to another fragment with backbone and/or more than one nucleotides. This make it a generally applicable modeling tool within DSSR.

I will post details on how the circular Z-DNA was generated in the next couple of days. We can polish the procedure together to better fit your needs, and hopefully it would be useful to other users as well.

Best regards,

Xiang-Jun

20

RNA structures (DSSR) / Re: Rebuilding circular Z-DNA

« Last post by Di_Liu on June 01, 2025, 09:08:25 pm »

Hi Xiang-Jun,

I checked your model, and I can confirm that it is what I wanted.

I'm also attaching the model I've built. The screenshot is a comparison of yours (blue) and mine (pink). I think we were using the G84 ring I've created, so our models can be overlayed. An extra step required for your model is to link the dinucleotide steps.

By the way, I think I figured out why I didn't get the email notification. I indeed received a notification on April 30, but I thought (mistakenly, probably due to the early morning...) I had already viewed the reply because you posted a reply on April 29, which was assumed to be the same message, so I didn't open a link for your April 30 message. I saw in the email that "More replies may be posted, but you won't receive any more notifications until you read the topic." I believe this setting is the reason I didn't the email notification of your follow-up message on May 5.

Best,

Di