Skip to main content

Advertisement

Table 2 Whole-genome comparisons between predictions and current Sc. pombe and S. cerevisiae annotations

From: CodingQuarry: highly accurate hidden Markov model gene prediction in fungal genomes using RNA-seq transcripts

  Nucleotide Exon Intron Gene
  Sn Sp Sn Sp Sn Sp Sn Sp
Sc. pombe (all 5124 genes)
CodingQuarry 98.6 98.9 90.3 89.4 92.6 95.2 87.5 83.0
AUGUSTUS 98.0 99.3 89.0 90.6 94.2 92.7 83.1 87.7
TransDecoder 93.4 99.2 80.8 85.4 85.3 96.6 76.3 72.5
S. cerevisiae (all 6575 genes)
CodingQuarry 97.2 99.5 76.1 87.2 64.4 65.8 76.6 88.3
AUGUSTUS 95.4 99.6 71.1 88.9 60.5 69.3 71.5 89.8
TransDecoder 87.8 99.7 67.1 75.0 60.5 70.1 67.8 68.0
  1. Sensitivity (Sn) is the proportion of a given feature (nucleotides/exons/introns/genes) in the annotation that are correctly predicted. Specificity (Sp) is the proportion of features in the predicted set that are correct. Sensitivity and specificity calculations for nucleotides are made on nucleotides within coding regions. Further descriptions of these measures are given in the Implementation subsection titled “Quantifying prediction results” The highest scores in each column for Sc. pombe and S. cerevisiae are shown in boldface.