Next: Acknowledgment Up: Discussion Previous: Future directions

Methods

Two programs have been created to achieve the desired color annotation. The first one is called bp-annotate. This program, annotates the PostScript structure plots that are created by the mfold package, or the PostScript secondary structure plots produced by the XRNA [Weiser & Noller, 1995] program. Either the bases are colored or the bases are replaced by colored dots. Both linear and double logarithmic scaling can be used. Two input files are needed; the unannotated PostScript file and an annotation file that we call the ann file because it ends with the suffix '.ann'. This annotation file can contain either P-num, H-num or probability information. A number of small auxiliary programs were created to generate this from different types of input information. The one most worth mentioning is called the boxct2ann program. It combines information from a probability boxplot produced, for example, by the Vienna package, and the ct file for the secondary structure to be annotated. The boxplot contains numbers, p_ij, for every possible base pair. These are the probabilities of designated base pairs. The ct file, or connection table file, is a common format for describing the bases and base pairs in an RNA secondary structure. The resulting ann file is a vector of probabilities, p_i, where i ranges over all bases of the sequence. For base pairs i.j, p_i = p_j = p_ij. If i is single stranded, then p_i is the probability that the base is single-stranded. In this case, p_i = 1 - Σ_j<i p_ji - Σ_j>i p_ij.

he second program, ss-annotate annotates the ss files that are used as input to the XRNA program. The figures can then be manipulated interactively by running the XRNA program. This is useful both in creating high quality plots, and for extracting detailed information about individual structural features from the annotated plots. The latter feature is particularly helpful in the analysis of predicted structure of very large sequences. The same color schemes and linear versus double logarithmic scaling are used in both annotation programs.

re comparison annotation is achieved through a program called ss-compare. It requires the ss files that are used as input to the program XRNA. In addition to an ss file, the program also requires the ct file for another reference structure on the same RNA. The base pairs in the ss file that are conserved in the reference structure are represented by thick lines that stand out from the non-conserved base pairs.

The programs bp-annotate and boxct2ann were written in Fortran. The programs ss-compare and ss-annotate were written in the C programming language. All of the software can be obtained from the first author and is available at http://mfold.rna.albany.edu/?q=mfold/download-mfold. Energy dot plots and annotated structure plots are also available for all structure predictions that are created with the mfold web server at http://mfold.rna.albany.edu/?q=mfold.

  Next: Acknowledgment Up: Discussion Previous: Future directions

Michael Zuker Institute for Biomedical Computing Washington University in St. Louis August 21 1998.