Skip to content

Conversation

MichielStock
Copy link
Owner

No description provided.

@MichielStock
Copy link
Owner Author

Just take a simple dynamic programming as a basis: https://www.cs.colostate.edu/~cs425/spring17/slides/Lecture_7_rna_folding.pdf

The algorithm is approximately as complex as Dijkstra.

@JietseV
Copy link

JietseV commented Jan 31, 2022

Hello
The feedback for your notebook:

Some typos:

  • In the explanation of Nussinov's algorithm underneath the formula there is written aglorithm
  • Second to last paragraph of the same block after S(i,j-1) your "If" is still in math mode
  • Try to be consistent with math mode, last line of second to last paragraph you have S(i,i-1) that is not written in math mode

In the basepair function:
You can return in your if else statement. Check your if statement, return True. The else is redundant, you always return false if the statement is not ok.

I would write your explanation of the functions and different steps at the top where you mention Nussinov's algorithm, but that is personal preference of course :)

Very clear explanation of the different parts, easy to follow! Also the visualization is very nice, pity there are no packages for RNA visualization, but nevertheless very clear representation of the structure!

@HeesooSong
Copy link

HeesooSong commented Jan 31, 2022

Hi,
I had a look into your notebook, and it was amazing to see that modeling the secondary structure of the RNA can be achieved through an optimization algorithm. The algorithm is nicely explained, and moreover, very interactive. It was fun to try out my own sequences. In addition to this, I would like to give some additional details for more perfection of your notebook :)

  1. Typo:
  • This was one of the first ever algorithms to be used in RNA folding but sadly this algorithm is often to (too) simple to get realistic results.
  1. Unit test block
    Although it became optional to add a separate folder for unit tests, it was still encouraged to add some unit tests in the notebook (make it a hidden cell). You can find more information about this in the instruction. I think it would be nice to test if the Nussinov matrix is correctly generated, for example by checking its format.

  2. Link
    You can simply generate a link button with markdown format. For instance:
    md"[TSSAR_structure](http://nibiru.tbi.univie.ac.at/forna/forna.html?id=url/name&sequence=GCUCGUGCGUCGAUGAACAACGCAGCUAGCUGCGAGAAUUAAUGUGAAUUGCAGGACACAUUGAUCAUCGACACUUCGAACGCACUUGCGGCCCCGGGUUCCUCCCGGGGCUACGCCUGUCUGAGCGUCGCUUGAAAAAAAAAAAAAAAAAAAAAA&structure=..(.(((((((((((..(...)))(.(.(.(.(((((.((((((((..((...)).)))))))))).))))))))).)).)))))(((((((((((((...)))..))))))..)).(.((.((((...)))))))))).................\)"
    This will create:
    TSSAR_structure

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants