The ITS2 database - Statistics Annotator

The statistics result from a comparison between annotations of all GenBank matching 'ITS2 OR "internal transcribed spacer 2"', the ITS-2DB and the HMM-Annotation. Databases were accessed at the 4h of June 2008. The HMM used for annotation is the eukaryote HMM (weakest HMM). For further information see Keller et al. Gene (submitted).

Figure 1: Distribution of positive and negative deviations of HMM-based annotations from GenBank or the ITS2-DB. Left: start annotation and right: end annotation. Plus-minus five nucleotides are in most cases acceptable as such a difference is small enough not to interfere with the structure prediction procedure. Positive deviations indicate that the use of the HMMs results in a larger ITS2 as compared with GenBank or the ITS2-DB, whereas negative values indicate the opposite. Numbers in parenthesis represent the sample size as the number of sequences with annotations by both HMMs and the respective database. Numbers below the boxplots are the total number of species with positive or negative deviation from the HMMs, respectively.

Figure 2: Density plots of Start deviations (alternate display to Figure 1)

Figure 3: Density plots of End deviations (alternate display to Figure 1)

Table 1: Number of ITS2 annotations obtained from three different sources (GenBank, ITS2-DB, and HMMs).

Table 2: Medians of positive differences (more nucleotides with HMM-based annotation), identical annotations, and negative differences (fewer nucleotides with HMM-based annotation) with their respective sample sizes in parenthesis resulting from comparisons of HMM-based annotations with GenBank and the ITS2-DB.

Table 3: Estimation of probabilities of erroneous annotations or undetected borders of 5.8S and 28S limits of the ITS2 by HMMs (Supplement).

If you use the database please cite.