Limits...
COFOLD: an RNA secondary structure prediction method that takes co-transcriptional folding into account.

Proctor JR, Meyer IM - Nucleic Acids Res. (2013)

Bottom Line: These aim to predict the most stable RNA structure.There exists by now ample experimental and theoretical evidence that the process of structure formation matters and that sequences in vivo fold while they are being transcribed.Here, we present a conceptually new method for predicting RNA secondary structure, called CoFold, that takes effects of co-transcriptional folding explicitly into account.

View Article: PubMed Central - PubMed

Affiliation: Centre for High-Throughput Biology, University of British Columbia, 2125 East Mall, Vancouver, BC, V6T 1Z4, Canada.

ABSTRACT
Existing state-of-the-art methods that take a single RNA sequence and predict the corresponding RNA secondary structure are thermodynamic methods. These aim to predict the most stable RNA structure. There exists by now ample experimental and theoretical evidence that the process of structure formation matters and that sequences in vivo fold while they are being transcribed. None of the thermodynamic methods, however, consider the process of structure formation. Here, we present a conceptually new method for predicting RNA secondary structure, called CoFold, that takes effects of co-transcriptional folding explicitly into account. Our method significantly improves the state-of-art in terms of prediction accuracy, especially for long sequences of >1000 nt in length.

Show MeSH

Related in: MedlinePlus

RNA secondary structures predicted by CoFold-A and RNAfold for the 23S rRNA of the γ-proteobacteria P. aeruginosa. The horizontal line corresponds to the RNA sequence of 2893-nt length. The structure predicted by RNAfold is shown above the horizontal line, and the one predicted by CoFold-A is shown below. Each arc corresponds to a base pair between the two corresponding positions along the sequence. Blue arcs correspond to correctly predicted base pairs (true positives), red arcs to incorrectly predicted base pairs (false positives) and black arcs to base pairs that are part of the reference structure, but missing from the prediction (false negatives). Orange arcs indicate base pairs of the reference structure that render it pseudo-knotted. Figure made with R-chie (37).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3643587&req=5

gkt174-F4: RNA secondary structures predicted by CoFold-A and RNAfold for the 23S rRNA of the γ-proteobacteria P. aeruginosa. The horizontal line corresponds to the RNA sequence of 2893-nt length. The structure predicted by RNAfold is shown above the horizontal line, and the one predicted by CoFold-A is shown below. Each arc corresponds to a base pair between the two corresponding positions along the sequence. Blue arcs correspond to correctly predicted base pairs (true positives), red arcs to incorrectly predicted base pairs (false positives) and black arcs to base pairs that are part of the reference structure, but missing from the prediction (false negatives). Orange arcs indicate base pairs of the reference structure that render it pseudo-knotted. Figure made with R-chie (37).

Mentions: The 23S rRNAs are the longest sequences of our data set with an average length of 3069 nt (min 2882 nt, max 3578 nt) and are thus some of the most challenging RNA structures to predict. Using CoFold and CoFold-A, we increase their prediction accuracy in terms of MCC w.r.t. RNAfold on average by 8 and 12%, respectively. Figure 4 shows, for the 23S rRNA of the γ-proteobacteria Pseudomonas aeruginosa, how the RNA structure predicted by CoFold-A compares with that predicted by RNAfold. The most apparent differences are that RNAfold predicts many incorrect mid- and long-range base pairs (red arcs spanning >100 nt), and that almost all of these disappear with CoFold-A. In addition, CoFold-A adds many correct mid- and long-range base pairs (blue arcs), see in particular those spanning almost the entire sequence. Overall, CoFold-A increases the MCC of RNAfold from 43 to 58%. This 15% rise in performance accuracy is due to a significant increase of the true positive rate () and an equally significant increase of the positive predictive value (). This is in line with is the typical behaviour seen for CoFold (Figure 2). The false positive rate for both prediction methods remains low at 0.01%.Figure 4.


COFOLD: an RNA secondary structure prediction method that takes co-transcriptional folding into account.

Proctor JR, Meyer IM - Nucleic Acids Res. (2013)

RNA secondary structures predicted by CoFold-A and RNAfold for the 23S rRNA of the γ-proteobacteria P. aeruginosa. The horizontal line corresponds to the RNA sequence of 2893-nt length. The structure predicted by RNAfold is shown above the horizontal line, and the one predicted by CoFold-A is shown below. Each arc corresponds to a base pair between the two corresponding positions along the sequence. Blue arcs correspond to correctly predicted base pairs (true positives), red arcs to incorrectly predicted base pairs (false positives) and black arcs to base pairs that are part of the reference structure, but missing from the prediction (false negatives). Orange arcs indicate base pairs of the reference structure that render it pseudo-knotted. Figure made with R-chie (37).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3643587&req=5

gkt174-F4: RNA secondary structures predicted by CoFold-A and RNAfold for the 23S rRNA of the γ-proteobacteria P. aeruginosa. The horizontal line corresponds to the RNA sequence of 2893-nt length. The structure predicted by RNAfold is shown above the horizontal line, and the one predicted by CoFold-A is shown below. Each arc corresponds to a base pair between the two corresponding positions along the sequence. Blue arcs correspond to correctly predicted base pairs (true positives), red arcs to incorrectly predicted base pairs (false positives) and black arcs to base pairs that are part of the reference structure, but missing from the prediction (false negatives). Orange arcs indicate base pairs of the reference structure that render it pseudo-knotted. Figure made with R-chie (37).
Mentions: The 23S rRNAs are the longest sequences of our data set with an average length of 3069 nt (min 2882 nt, max 3578 nt) and are thus some of the most challenging RNA structures to predict. Using CoFold and CoFold-A, we increase their prediction accuracy in terms of MCC w.r.t. RNAfold on average by 8 and 12%, respectively. Figure 4 shows, for the 23S rRNA of the γ-proteobacteria Pseudomonas aeruginosa, how the RNA structure predicted by CoFold-A compares with that predicted by RNAfold. The most apparent differences are that RNAfold predicts many incorrect mid- and long-range base pairs (red arcs spanning >100 nt), and that almost all of these disappear with CoFold-A. In addition, CoFold-A adds many correct mid- and long-range base pairs (blue arcs), see in particular those spanning almost the entire sequence. Overall, CoFold-A increases the MCC of RNAfold from 43 to 58%. This 15% rise in performance accuracy is due to a significant increase of the true positive rate () and an equally significant increase of the positive predictive value (). This is in line with is the typical behaviour seen for CoFold (Figure 2). The false positive rate for both prediction methods remains low at 0.01%.Figure 4.

Bottom Line: These aim to predict the most stable RNA structure.There exists by now ample experimental and theoretical evidence that the process of structure formation matters and that sequences in vivo fold while they are being transcribed.Here, we present a conceptually new method for predicting RNA secondary structure, called CoFold, that takes effects of co-transcriptional folding explicitly into account.

View Article: PubMed Central - PubMed

Affiliation: Centre for High-Throughput Biology, University of British Columbia, 2125 East Mall, Vancouver, BC, V6T 1Z4, Canada.

ABSTRACT
Existing state-of-the-art methods that take a single RNA sequence and predict the corresponding RNA secondary structure are thermodynamic methods. These aim to predict the most stable RNA structure. There exists by now ample experimental and theoretical evidence that the process of structure formation matters and that sequences in vivo fold while they are being transcribed. None of the thermodynamic methods, however, consider the process of structure formation. Here, we present a conceptually new method for predicting RNA secondary structure, called CoFold, that takes effects of co-transcriptional folding explicitly into account. Our method significantly improves the state-of-art in terms of prediction accuracy, especially for long sequences of >1000 nt in length.

Show MeSH
Related in: MedlinePlus