Limitations of GENSCAN
-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .-. .
||X|||\ /|||X|||\ /|||X|||\ /|||X|||\ /|||X|||\ /|||X|||\ /|||X|||\ /|||X|||\ /|||X|||\ /|
|/ \|||X|||/ \|||X|||/ \|||X|||/ \|||X|||/ \|||X|||/ \|||X|||/ \|||X|||/ \|||X|||/ \|||X||
' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-' `-
GENE NUMBER.
The number of genes predicted in
a sequence is usually approximately correct but it may also
happen that, for instance, a predicted gene splices together exons
from two real genes, or vice versa.
ORGANISM.
The program was designed primarily to predict
genes in human/vertebrate genomic sequences: accuracy
may be lower for non-vertebrates.
In particular, the vertebrate version of the program performs
fairly well on Drosophila sequences and the
maize and Arabidopsis versions perform fairly well
on their respective organisms, but the program version for
C. elegans sequences is still in the experimental stages.
Other organisms have not been systematically tested.
For prokaryotic or yeast sequences, the programs
Glimmer,
developed by Salzberg and colleagues, and/or
GeneMark,
developed by Borodovsky and McIninch, are recommended.
BIASES IN TEST SET.
The Burset/Guigó test set is undoubtedly biased toward
short genes with relatively simple exon/intron structure:
as a consequence, the statistics given in the tables on the
accuracy page
may not be completely representative of the
performance of the programs on typical genomic DNA sequences.
EXON/FEATURE TYPE.
As a rule, internal exons are predicted more accurately
than initial or terminal exons, and exons are predicted
more accurately than polyadenylation or promoter signals.
Predicted promoters, in particular, are not reliable.
If promoters are of primary interest, you may wish to
try Martin Reese's
NNPP program.
PLANT SPLICE SIGNALS.
If identification of splice sites in plant pre-mRNAs is of
primary interest, the
SplicePredictor
program developed by Brendel and colleagues is recommended.
Back to the GENSCAN web server.
Address any comments/questions/suggestions to:
cburge@mit.edu