Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

RNAmountAlign: Efficient software for local, global, semiglobal pairwise and multiple RNA sequence/structure alignment

Fig 5

Relative frequency histograms of alignment scores for local (left), semiglobal (middle) and global (right) alignments of random RNAs produced by RNAmountAlign.

For the 5S rRNA AY544430.1:375-465 from the Rfam 12.0 database having nucleotide relative frequencies pA = 0.25, pC = 0.27, pG = 0.26, pU = 0.21, we generated 10,000 random sequences having the same nucleotide relative frequencies, each of length 400 nt. For each method (local, semiglobal, global), RNAmountAlign was run using default parameters to determine the optimal pairwise alignment score, when aligning the 5S rRNA with each random RNA, thus producing relative frequency histograms which were subsequently fit by the normal distribution (ND), extreme value distribution (EVD) and gamma distribution (GD). As expected by Karlin-Altschul theory [12], local alignment scores are best fit by EVD, while semiglobal alignment scores are best fit by ND. Our conclusions of the best fitting distributions were additionally supported by d computations of variation distance, symmetrized Kullback-Leibler distance, and χ2 goodness-of-fit tests (data not shown).

Fig 5

doi: https://doi.org/10.1371/journal.pone.0227177.g005