Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2009 Aug 28;5(8):e1000483. doi: 10.1371/journal.pcbi.1000483

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

Bradley, Holmes.

This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.

PMC Copyright notice

We simulated the evolution of three structural RNAs under the TKFST model. The simulation included two sister species at unit distance from the ancestral sequence, plus one “outgroup” whose branch length was varied between by selecting 25 equally spaced values of in this range, spaced apart. We then simulated 25 alignments for each value of , using TKFST model parameters described in the text. The Alignment Metric Accuracy (AMA) is, roughly, the proportion of residues that are correctly aligned, averaged over all pairs of sequences (see [54] for a precise definition; we set the AMA Gap Factor to 1). The AMA between the true alignment and the inferred alignment was measured for various statistical alignment inference procedures. These procedures are described in the text, but may be summarized very briefly as ML under the true model (“Indiegram”); greedy approximate-ML progressive alignment by single-linkage clustering with pair SCFGs (“Stemloc”); sequence annealing, a form of posterior decoding to maximize a sum-over-pairs accuracy metric, using pair SCFGs to get the posterior probabilities (“Stemloc-AMA”); statistical alignment using the TKF91 model, i.e. linear gap-penalties (“TKF91”); and statistical alignment using a long-indel model, i.e. affine gap-penalties (“Long Indel”).

Inline graphic — We simulated the evolution of three structural RNAs under the TKFST model. The simulation included two sister species at unit distance from the ancestral sequence, plus one “outgroup” whose branch length was varied between by selecting 25 equally spaced values of in this range, spaced apart. We then simulated 25 alignments for each value of , using TKFST model parameters described in the text. The Alignment Metric Accuracy (AMA) is, roughly, the proportion of residues that are correctly aligned, averaged over all pairs of sequences (see [54] for a precise definition; we set the AMA Gap Factor to 1). The AMA between the true alignment and the inferred alignment was measured for various statistical alignment inference procedures. These procedures are described in the text, but may be summarized very briefly as ML under the true model (“Indiegram”); greedy approximate-ML progressive alignment by single-linkage clustering with pair SCFGs (“Stemloc”); sequence annealing, a form of posterior decoding to maximize a sum-over-pairs accuracy metric, using pair SCFGs to get the posterior probabilities (“Stemloc-AMA”); statistical alignment using the TKF91 model, i.e. linear gap-penalties (“TKF91”); and statistical alignment using a long-indel model, i.e. affine gap-penalties (“Long Indel”).