Training data sets
| Training data sets | Exonomy | GeneSplicer | GlimmerM | Unveil |
| Arabidopsis.thaliana.tar.gz | Yes | Yes | Yes | Yes |
| Aspergillus.fumigatus.tar.gz | Yes | - | Yes | Yes |
| Aspergillus.spp.tar.gz | Yes | - | Yes | Yes |
| Homo.sapiens.tar.gz | Yes | Yes | - | Yes |
| Mus.musculus.tar.gz | Yes | - | - | Yes |
| Plasmodium.falciparum.tar.gz | - | Yes | Yes | - |
| chrom-coords-build31.tar.gz : (Human). Chromosome coordinates of 7709 high-confidence CDSs obtained through EnsMart with the generous help of Ewan Birney. GFF format. Coordinates are relative to Human build 31 chromosome files. | Yes | - | - | Yes |
| train.tar.gz : (Human). One multi-FASTA file containing 1500 high-confidence CDSs from EnsMart plus 2000bp on each side; also, one GFF file containing exon coordinates for all 1500 CDSs within those sequences. | Yes | - | - | Yes |
| test.tar.gz :(Human). 483 FASTA files and 483 GFF files containing coordinates of CDSs and their exons in those fasta files. Disjoint set from train.tar.gz. | Yes | - | - | Yes |