findpair -p and -z options

[auffinger]

Dear Xiang-Jun,

Recently, we tried to use find_pair with the -p option that gives more complete listings of base pairs. Yet, the output of find_pair with the -p option seems not comptatible with analyze. Hence we tried to convert the extended output of findpair into the simplier one, without any clue about where to get the right info about the (|, x, and +) signs associated with the (0, 1, 9) numbers (helix breaks ?). is there any easier way, to make analyze work on the complex output of find_pair ?

Also,

    1    1  197  344 ....>C:...1_:[.DT]T-**+-A[.DA]:..21_:D<....
             T+A     -0.64    3.35    1.34  -23.31    0.25  -70.84
             T+A [2]  N3 - N7  2.76  O4 - N6  2.82
                    parallel cis cis
            +++     69.2     74.1     23.3     73.2   -109.2   -127.6
         3.66    1.34   23.31    6.65    4.51   -9.93   53.38   21.36

what do the three +++ mean here as well as other undocumented thinks.

Thanks for help,

Kind regards,

Pascal

[xiangjun]

Hi Pascal,

The "-p" option of find_pair is for pairwise checking of all possible base-pairs, and for identifying higher-order base associations as demonstrated in the 2008 3DNA NP paper (Recipe no. 5). It was not intended to be used with analyze, as I initially added the option there. The "-z" option, on the other hand, was mostly used to decide suitable criteria of which base-pair should be included in a helical region. Its output information is far too technical for outside users.

Regarding the output parameters from the "-p" option, the header section should be of some help:
[pre:byvny7l4]Six-line information for each base-pair as follows:
   #1: Overall serial number, local serial number, paired residue numbers,
       detailed pairing residue information.
   #2: One-letter base-pair followed by six base-pair parameters (shear,
       stretch, stagger, buckle, propeller, opening). The parameters are
       with respect to the Watson-Crick base reference frame. There are
       two types of base-pair orientation: M-N means the two bases have
       opposite orientations as in Watson-Crick base-pair; M+N means the
       two bases have the same local orientations as in Hoogsteen base-
       pair. All possible base pairing patterns can then be classified
       based on the six parameters, among which shear, stretch and opening
       are most discriminative.
   #3: H-bonding information (atom pair followed by their distance).
   #4: Overall classification of the base-pair (anti-parallel vs parallel
       based on relative z-axis of the two bases, cis vs trans based on
       x-axis and C1-RN9/YN1 directions).
   #5: Relative directions of the three axes and their numerical values.
       The last 3 numbers are the angles between the glycosidic bonds, and
       the two chi torsion angles.
   #6: The actual parameters used to locate the base-pair in question.[/pre:byvny7l4]
Some of the undocumented features of 3DNA are experimental or are related to still unpublished work. I am hoping to be able to devote more efforts to research on nucleic acid structures, but it all depends.

Xiang-Jun
[hr:byvny7l4][/hr:byvny7l4]
PS. BTW, isn't FR3D sufficient for tasks related to the identification and classification of base-pairs? The FR3D webpage contains tons of information; some of which you may find helpful.

[mauricio esguerra]

Hi,

Perhaps you are already aware of Yurong Xing's database on RNA base-pairs?, if not, it might be useful to you.
You can find it at:

http://http://bps.rutgers.edu

You can search by PDBID, or NDBID if you go to:

http://http://bps.rutgers.edu/search/strid

You can find more information on what the output means, specially the helical region classification at:

http://http://bps.rutgers.edu/bps/help_bp

Hope this helps,

Mauricio

[auffinger]

Dear Mauricio,

Thanks your repy, but yes, I am aware of this great BPS site and hope you are also aware about our http://tatooine.u-strasbg.fr/~sws/SwS.html site that is currently undergoing a big restructuration. I am still surprised to find out how much there is still to say about base pairs.

Cheers,

Pascal