Table I.

Evaluation of current gene annotation in Arabidopsis and rice

Annotation CategoryaAnnotation Support at AtGDBAnnotation Support at OsGDB
TotalCoveragebTotalCoverageb
=0%<50%<100%=100%=0%<50%<100%=100%
Nonspliced5,3311,6592666582,74812,6338,1445121,0892,888
Spliced
All introns confirmed18,4191123,95814,34912,7581253,4329,201
≥75% intron confirmation839276471651,575511,173351
≥50% intron confirmation1,047233716982,5124201,832260
>0% intron confirmation1,327965342202,4961,43596794
No introns confirmed5,0083,9506752958830,14725,2252,9421,604376
Annotation totals31,9715,6092,2786,61617,46862,12133,3695,48510,09713,170
• a Annotations with multi-exon structural definitions (spliced annotations) are categorized by the ratio of introns confirmed by EST and/or cDNA spliced alignment to the total number of predicted introns. For example, an annotation predicting a gene model with five exons would by definition have four introns. If three of these intron positions are confirmed, this annotation would be placed in the ≥75% category. If only two were confirmed, the annotation would fall into the ≥50% category.

• b Annotations are further categorized by coverage. Coverage percentages denote the fraction of the annotation-defined exon regions overlapped by a sequence alignment (the fraction represented by physical sequence in the form of EST and/or cDNA).