plot of basepairing probabilities

[tejajo]

Hi,
I want to generate a RNAfold type dotplot of my base-match
probabilities for pair-wise alignment of RNAs. The probabilities could
be color coded or sizes of rectangles representing probabilities
should vary as in the case of RNAfold's dotplot.
Any ideas on how I could generate this using MATLAB, R or any other
tool for that matter ?
Thanks in advance..
Tejal

[Ana Kozomara]

Hi Tejajo.
To reproduct RNAfold dotplot you can use Ct Format representation to code you base pairings: in the same manner as RNAfold does.
Every RNAfold output is followed by one Vienna format file (brackets notation) and one Ct Format  file.

Exemple:

UCGGAUCCGAUUCGGCCAUGAAUGACUCCGA
(((((((..(((((....))))))).)))))

Corresponding Ct file:

    1 U       0    2   31    1
    2 C       1    3   30    2
    3 G       2    4   29    3

    4 G       3    5   28    4
    5 A       4    6   27    5
    6 U       5    7   25    6
    7 C       6    8   24    7
    8 C       7    9    0    8
    9 G       8   10    0    9
   10 A       9   11   23   10

   11 U      10   12   22   11
   12 U      11   13   21   12
   13 C      12   14   20   13
   14 G      13   15   19   14
   15 G      14   16    0   15
   16 C      15   17    0   16
   17 C      16   18    0   17

   18 A      17   19    0   18
   19 U      18   20   14   19
   20 G      19   21   13   20
   21 A      20   22   12   21
   22 A      21   23   11   22
   23 U      22   24   10   23
   24 G      23   25    7   24

   25 A      24   26    6   25
   26 C      25   27    0   26
   27 U      26   28    5   27
   28 C      27   29    4   28
   29 C      28   30    3   29
   30 G      29   31    2   30
   31 A      30    0    1   31

I invite you to take a look (in case you are not familiar with) at:
http://rna.tbi.univie.ac.at/help.html
And also to try, where you can find Ct format outputs:
http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi


Hope this helps,
Ana

[tejajo]

Hi Ana,
Thanks for your reply. I am sorry if I haven't understood your idea. I
feel that Ct file doesn't take any probability values, but only
possible base-match probabilities, as in the form of constraints.
However, what I have with me is base-match probabilities (probability
of having an alignment edge at given base positions, each from a
sequence). This means, I have two sequences (not necessarily the
same), their corresponding base-match probabilities generated using
partition function in the following format , for sequences A and B:

posA posB BP_prob
1 1 0.87
1 2 0.45
1 5 0.3
3 3 0.1
3 4 0.24

Can we incorporate probabilities into Ct file in order to generate a
dot-plot ?

cheers,
Tejal