Msa Score For Rna

Question

[User 0431]

how to calculate the scores without using references alignment or structure? anyone have the good function to do that??

that mean:

I generate many RNA alignments in each iterative for the same family to find the optimal one using optimization algorithm:

in the algorithm it depend on objective function(score1) to find the best alignment

family1

1. alignmt1 score1 score2

2. Alignmt2 score1 score2

3. Alignmt3 score1 score2

4. Alignmt4 score1 score2

5. Alignmt5 score1 score2

score1 is the objective function without using references,or structure like using the(Sum-of-pairs,NorMD)

score2 is the qulatiy of the aligment with refernces like using the(Qscore)

in my optimization program it depend on the score1 function to select the best alignment

I assume there is correlation between score1 and score2???

I want to know if the MSA actually better fits the reality or not and it should be correlation there between score1 and score2. Am I right or not? if not what I should used as objective function which can fits the reality, then the MSA can fits the reality?

[Lars Juhl Jensen]

I think you need to explain better what you have done and what you are trying to do. I am at least unable to decipher it from what you have written.

[User 0431]

please any one can answer me

[User 0431]

I am asking if there is a correlation between score1 and score2? which score1 is objective function like sum-of-pairs or norMD and score2 is to measure the alignmnet quality like Qscore, SPS,TC.

Answer 1

[Andreas]

Not entirely sure if I understand you correctly, but one way to measure the alignment quality of non-coding RNAs without a reference is the so called 'structure conservation index' introduced by RNAz (Stefan Washietl).

It was also used for benchmarking RNA alignment programs (BRAlibase; see PMID 15860779 and PMID 17062125). A program to compute this score (scif) can be found on the BRAliBase 2.1 website.

There should be lots of other ways to do this. Just check how some of the non-coding RNA alignment programs work (like MASTR etc.)

Andreas

[User 0431]

measure the alignment quality of non-coding RNAs without a reference and without structure.. the measurement function that may fit the real alignmnet

[Andreas]

The SCI does not need a reference alignment or structure

[User 0431]

what is SCI stand for??

[Andreas]

SCI = Structural Conservation Index

[User 0431]

Dear sir than you for your answer, I wanna to check if the sum-of-pairs is truly.. means if i am optimize the alignment by using objective function (sum-of-pair) to find the good score with out using secondary or references here ,should that alignment is good when validate it with the reference alignment(from banchmark).