Identification of multiplets in DSSR

In DSSR (and find_pair -p from the original 3DNA suite), multiplets is defined as “three or more bases associated in a coplanar geometry via a network of hydrogen-bonding interactions. Multiplets are identified through inter-connected base pairs, filtered by pair-wise stacking interactions and vertical separations to ensure overall coplanarity.”

DSSR detects multiplets automatically, and outputs a corresponding MODEL/ENDMDL delineated PDB file (dssr-multiplets.pdb by default) where each multiplet is laid in the most extended view in terms of base planes. The DSSR Nucleic Acids Research (NAR) paper contains four examples (in supplemental Figures 1, 3, 4, and 7) to illustrate this functionality. Please refer to Reproducing results published in the DSSR-NAR paper on the 3DNA Forum for details.

Recently, I read the article titled InterRNA: a database of base interactions in RNA structures by Appasamy et al. in NAR. In Figure 2 (linked below) of the paper, the authors showcased a sextuple (hexaplet) identified in the E. coli ribosome (PDB id: 4tpe), along with six base-base H-bonds contained therein.

Hexaplet GUUAAA in 4tpe
Figure 2. Example of the user interface displaying an InterRNA database record.

With interest, I tried to run DSSR on the PDB entry 4tpe. As it turns out, ‘4tpe’ has been merged into 4u27 in mmCIF format. I ran DSSR (v1.4.6-2015dec16) in its default settings on ‘4u27’ and get the following summary of results.

# x3dna-dssr -i=4u27.cif -o=4u27.out
    total number of base pairs: 4822
    total number of multiplets: 680
    total number of helices: 264
    total number of stems: 566
    total number of isolated WC/wobble pairs: 193
    total number of atom-base capping interactions: 615
    total number of hairpin loops: 215
    total number of bulges: 137
    total number of internal loops: 244
    total number of junctions: 108
    total number of non-loop single-stranded segments: 83
    total number of kissing loops: 14
    total number of A-minor (type I and II) motifs: 246
    total number of ribose zippers: 127
    total number of kink turns: 15

Among the 680 DSSR-identified multiplets, two hexaplets (one on chain “AA”, and another on “CA”) match those reported by Appasamy et al., as shown below:

 678 nts=6 GUUAAA 1:AA.G404,1:AA.U438,1:AA.U439,1:AA.A496,1:AA.A498,1:AA.A499
 679 nts=6 GUUAAA 1:CA.G404,1:CA.U438,1:CA.U439,1:CA.A496,1:CA.A498,1:CA.A499

For illustration, the hexaplet #678 is extracted from dssr-multiplets.pdb to file 4u27-hexaplet.pdb (download the coordinates) and shown below. The figure is generated by DSSR and PyMOL, as detailed in Reproducing results published in the DSSR-NAR paper on the 3DNA Forum.

x3dna-dssr -i=4u27-hexaplet.pdb -o=4u27-hexaplet.pml --hbfile-pymol 

Hexaplet GUUAAA in 4u27
DSSR-identified hexaplet GUUAAA in 4u27.

DSSR identifies 6 base pairs in the hexaplet:

# x3dna-dssr -i=4u27-hexaplet.pdb --idstr=short
List of 6 base pairs
      nt1            nt2           bp  name        Saenger    LW  DSSR
   1 G404           A498           G+A --          n/a       tSS  tm+m
   2 G404           A499           G+A --          n/a       cWH  cW+M
   3 U438           A496           U-A rHoogsteen  24-XXIV   tWH  tW-M
   4 U439           A496           U-A --          n/a       cH.  cM-.
   5 U439           A498           U-A WC          20-XX     cWW  cW-W
   6 A496           A498           A+A --          n/a       cWH  cW+M

It detects a total of 9 H-bonds as shown below. In addition to the 6 base-base H-bonds noted by Appasamy et al., DSSR also finds 3 sugar-base H-bonds (#1, #2, and #4, labeled in green) that obviously play a role in stabilizing the high-order base association.

# x3dna-dssr -i=4u27-hexaplet.pdb --get-hbonds --idstr=short
   11    59  #1     o    3.017 O:N O2'@G404 N3@U439
   11   104  #2     o    2.578 O:N O2'@G404 N1@A498
   18   125  #3     p    3.089 O:N O6@G404 N6@A499
   21    96  #4     o    3.289 N:O N2@G404 O2'@A498
   21   106  #5     p    2.797 N:N N2@G404 N3@A498
   39    78  #6     p    2.944 N:N N3@U438 N7@A496
   61    81  #7     p    3.167 O:N O4@U439 N6@A496
   61   103  #8     p    2.662 O:N O4@U439 N6@A498
   82   103  #9     p    3.152 N:N N1@A496 N6@A498




Thank you for printing this article from Please do not forget to visit back for more 3DNA-related information. — Xiang-Jun Lu