Query in 3DNA-DSSR output vs W3DNA output

[sankaran]

Hi,
I am a Postgraduate student currently pursuing a project. For my project work, I am using 3DNA-DSSR (v1.9.10-2020apr23.) to calculate the DNA shape parameters. I am writing to you to get clarity on the output from the x3DNA-DSSR and the W3DNA. The base step parameters from the 3DNA-DSSR and W3DNA are different. In the DSSR, the step parameters (Shift, Slide, Rise, Tilt, Roll, Twist) for the last base pair is given as 9999 whereas, in the output from the web server, the step parameters for the first base pair is given as 0.000. Considering the fact that the base pair step parameters are calculated based on the current and previous base pairs, 0.00 for the first base pasir would be ideal. but in the case of DSSR, the first base pair gets the value. Can you please let me know why is there a difference? I have attached the output files along with the PDB.
Thank you

[xiangjun]

Hi,

Thanks for using DSSR and 3DNA, and for posting your question on the 3DNA Forum.

For a duplex with N base pairs, there are N-1 base-pair steps. In the 3DNA suite of programs, the analyze program produces a file with content like below (using your attached example):

   8 # base-pairs
   0 # ***local base-pair & step parameters***
#        Shear    Stretch   Stagger   Buckle   Prop-Tw   Opening     Shift     Slide     Rise      Tilt      Roll      Twist
G-C     -0.270    -0.132    -0.197    -0.052     1.100     4.414     0.000     0.000     0.000     0.000     0.000     0.000
A-T     -0.499    -0.435     0.176     1.262   -10.089   -12.655    -1.694    -0.607     3.330    -2.903    -4.465    34.601
G-C     -0.115     0.069     0.407     5.730    -4.240    10.116     0.638    -0.387     3.180    -0.622     0.059    39.925
G-C     -0.876    -0.275     0.220     8.648    -3.894    -3.932    -0.713     0.224     3.224    -0.525     4.756    27.462
C-G      0.513    -0.078     0.087    12.122   -13.041    -1.249     0.731    -0.167     3.314     1.351     2.611    38.586
T-A     -0.066    -0.076     0.266   -17.997    13.206    -0.182     2.088     2.132     7.857   -17.640     5.211    44.686
T-T      1.892    -1.893     0.406     0.252     7.615     6.070     1.058     0.614     3.339    -4.474    11.137    38.739
A-A     -4.400     1.650     0.278   -10.936     4.849  -115.147    -8.924    -0.354     6.529   -34.317    10.409    20.463

The 3DNA rebuild program can then read this parameter file and build a model accordingly. Here the six parameters (highlighted in red) along with the first base pair are just space fillers. ANY numeric values will serve the purpose.

Now in DSSR, I have changed the format as below:

# 7 (no. of base pairs)
#bp      Shear     Stretch    Stagger    Buckle   Propeller   Opening      Shift      Slide      Rise       Tilt       Roll       Twist
G-C    -0.2701    -0.1317    -0.1971    -0.0521     1.0996     4.4143    -1.6945    -0.6073     3.3302    -2.9026    -4.4649    34.6011
A-T    -0.4989    -0.4352     0.1758     1.2619   -10.0888   -12.6549     0.6376    -0.3870     3.1803    -0.6220     0.0588    39.9253
G-C    -0.1149     0.0686     0.4070     5.7305    -4.2403    10.1156    -0.7126     0.2239     3.2238    -0.5251     4.7555    27.4624
G-C    -0.8762    -0.2749     0.2201     8.6484    -3.8937    -3.9324     0.7312    -0.1669     3.3143     1.3511     2.6115    38.5859
C-G     0.5131    -0.0780     0.0868    12.1222   -13.0413    -1.2495     2.0883     2.1322     7.8572   -17.6400     5.2115    44.6864
T-A    -0.0657    -0.0764     0.2658   -17.9967    13.2060    -0.1819     1.0584     0.6138     3.3386    -4.4736    11.1373    38.7389
T-T     1.8918    -1.8929     0.4065     0.2517     7.6152     6.0696     999999     999999     999999     999999     999999     999999

The number 999999 in DSSR makes the space-filling purpose of the six extra step parameters more obvious than 0.000 in 3DNA. They are put into the line with the final base pair as I feel this arrangement more natural. Most importantly, the DSSR output is intended to be fed into a modeling module of DSSR Pro, not to be used with the original 3DNA rebuild program.

The --analyze option has been removed from DSSR as of version 2.0 to avoid the confusion you experienced. Thus DSSR basic does not have this feature any more, whilst DSSR Pro has a new, much enhanced module in its place. DSSR Pro has completely superseded 3DNA, with a streamlined user interface and many advanced features (especially in modeling).

Best regards,

Xiang-Jun