Computational Inference of Homologous Gene Structures in the Human Genome

Table 3.

Comparison of GenomeScan-Predicted Genes on Human Chromosome 22 with Annotated Genes

Variable Category of gene
Known Related Pseudo Other
Total no. of genes annotated in Chr22 307 120 132 245
Percent of annotated genes in category covered by GenomeScan-predicted genes 95 88 67 67
Percent of annotated genes in category overlapping multiple GenomeScan-predicted genes 9 6 1 1
Percent of annotated exons in category covered by GenomeScan-predicted exons 94 92 74 74
Percent of allGenomeScan-predicted genes which match annotated genes in category 51 15 11 12
  • Genes were predicted with GenomeScan in the masked May 19, 2000 version of the Chr22 sequence, and compared to the Sanger Centre annotation (http://www.sanger.ac.uk/HGP/Chr22/cwa_archive/Release_2_19-05-2000.shtml). Known genes, related genes, predicted genes, immunoglobulin “gene segments”, and pseudogenes are distinguished in the annotation; the “Other” category includes annotated predicted genes and gene segments. GenomeScan predicted a total of 648 genes in the Chr22 sequence, of which 11% did not overlap any annotated gene; thus, the entries in the last row of the table total 89% rather than 100%.

This Article

  1. Genome Res. 11: 803-816

Preprint Server