Netiquette · Download · News · Gallery · Homepage · DSSR · Web-DSSR · DSSR Manual · Reproduce DSSR · DSSR-Jmol · DSSR-PyMOL · Web-SNAP

Author Topic: A question about base inclination  (Read 1169 times)

Offline lvelve0901

  • regular
  • *
  • Posts: 34
    • View Profile
A question about base inclination
« on: August 24, 2016, 09:28:49 am »
Hi, xiangjun,

I read a paper about A-tract DNA and it seemed that they reported base inclination for two base in a base pair respectively.

(MacDonald, D., et al. (2001). "Solution structure of an A-tract DNA bend." J Mol Biol 306(5): 1081-1098.)

But in DSSR there is only one inclination value in helical parameters. I am wondering are they the same definition?

Thank you so much for your help.

Best,
Honglue

Offline xiangjun

  • Administrator
  • regular
  • *****
  • Posts: 1184
    • View Profile
    • 3DNA homepage
Re: A question about base inclination
« Reply #1 on: August 24, 2016, 10:56:07 am »
Hi Honglue,

I browsed through the MacDonald et al. JMB paper quickly, and noticed that the helical parameters were calculated with FREEHELIX (Table 1). The NDB maintains a copy of freehelix98, if you want to get into the details.

In 3DNA, the helical parameters (including tip/inclination) are calculated for double helices, by default. However, a similar set of parameters is also available for each strand. For example, the following command:

find_pair -s 355d.pdb | analyze

It takes the classic Dickerson B-DNA dodecamer as a "pseudo" continuous single-stranded structure, and outputs (in file "355d.outs") helical parameters for each consecutive dinucleotide. Moreover, the auxiliary files "bp_step.par" and "bp_helical.par" (a bit misnomer with the bp_ prefix) can be used for exact rebuilding, just as for a duplex. See the $X3DNA/doc/tech-details.pdf for how the parameters are calculated. 3DNA uses exactly the same definition for all such parameters, either in a duplex or a single-stranded structure.

DSSR follows the same algorithm as 3DNA (in analyze/rebuild) for calculating these parameters. However, the (main) output file contains helical parameters for the duplexes only. DSSR already computes an overwhelming number of parameters, and I've tried to make the default (main) output 'simple'. If you are interested in such single-stranded helical parameter from DSSR, a natural place is a new section in the "dssr-torsions.txt" file. What do you think?

HTH,

Xiang-Jun
Dr. Xiang-Jun Lu [律祥俊]
Email: xiangjun@x3dna.org
Homepage: http://x3dna.org/
Forum: http://forum.x3dna.org/

Offline lvelve0901

  • regular
  • *
  • Posts: 34
    • View Profile
Re: A question about base inclination
« Reply #2 on: August 24, 2016, 12:13:52 pm »
Hi, xiangjun,

That is very helpful! Thank you so much. I am just curious about the different definitions but I think I don't need the single strand parameters for now.

Actually, I am analyzing the structure (1fzx.pdb) in that paper using both Curves+ and x3dna-dssr. To my surprise, the value of the base pair parameters and helical parameters are slightly different. Is it arise from different definitions?

Also, for some helical parameters (e.g x-displacement, tip, inclination etc.), Curves+ will give you values for all 12 base pairs (see the section (A) BP-Axis in 1fzx_0001_out.lis file). But it seemed that x3dna-dssr only give you 11 values in the helical parameters section (it miss one terminal DG12-DC13 in the attached figure). Could you explain the difference to me? Is it due to different definitions?

I also attach the pdb file.

Thank you so much.

Best,
Honglue

Offline xiangjun

  • Administrator
  • regular
  • *****
  • Posts: 1184
    • View Profile
    • 3DNA homepage
Re: A question about base inclination
« Reply #3 on: August 24, 2016, 12:43:15 pm »
Quote
Actually, I am analyzing the structure (1fzx.pdb) in that paper using both Curves+ and x3dna-dssr. To my surprise, the value of the base pair parameters and helical parameters are slightly different. Is it arise from different definitions?

Slight differences between DSSR and Curves+ in base-pair and helical parameters are expected, even for Watson-Crick pairs (for non-canonical pairs, the differences would be immediately obvious). Even within Curves+ itself, setting fit or not will also make a slight difference.

DSSR and Curves+ all use the standard base reference frame, and give quite similar (but not identical) parameters for structures consisting of Watson-Crick pairs. See my blog post Curves+ vs 3DNA dated 2012-01-10. The two are subtly different in the implementation of the standard base reference, and the definition of the parameters. These details may well be out of general interest, I guess. For those who do, check the source code.

Quote
Also, for some helical parameters (e.g x-displacement, tip, inclination etc.), Curves+ will give you values for all 12 base pairs (see the section (A) BP-Axis in 1fzx_0001_out.lis file). But it seemed that x3dna-dssr only give you 11 values in the helical parameters section (it miss one terminal DG12-DC13 in the attached figure). Could you explain the difference to me? Is it due to different definitions?

The question of one-less bp helical parameters has been asked in the 3DNA Forum before. Here is an except from my reply to a similar question dated August 22, 2007:

Quote
The number of local helical parameters should be ONE less than the number of base-pairs.

Note that helical parameters refer to a double helix region, and the smallest helix fragment is a dinucleotide step, which is one less than the number of bps. In 3DNA, the helical parameters for each dinucleotide step are defined symmetrically such that they are exactly the same for the two bps.

HTH,

Xiang-Jun
Dr. Xiang-Jun Lu [律祥俊]
Email: xiangjun@x3dna.org
Homepage: http://x3dna.org/
Forum: http://forum.x3dna.org/

Offline lvelve0901

  • regular
  • *
  • Posts: 34
    • View Profile
Re: A question about base inclination
« Reply #4 on: August 24, 2016, 01:07:15 pm »
Hi, xiangjun,

Thank you so much! You are so helpful.

I will keep exploring DSSR in my project.

Best,
Honglue

 

Created and maintained by Dr. Xiang-Jun Lu[律祥俊]· Supported by the NIH grant R01GM096889 · Dr. Lu is currently a member of the Bussemaker Laboratory at the Department of Biological Sciences, Columbia University. The project is in collabration with the Olson Laborarory at Rutgers where 3DNA got started.