The Baylor College Of Medicine Computational Biology Group
Houston, TX
announces a new service
HEXON
Prediction of internal exons in Human DNA sequences
NOTE: This service is temporarily being provided through the
University of Houston Gene-Server. Only two jobs will be run at a
time.
Analysis of uncharacterized human sequences is available by sending a
file containing a sequence name and a sequence (no more than 80
characters/line) to:
service at theory.bchs.uh.edu
with the subject line "HEXON".
Example: mail -s HEXON service at theory.bchs.uh.edu < test.seq
where test.seq a file with the sequence.
Method description:
**********************
The method is based on the discriminant analysis of open reading
frames flanked by GT and AG base pairs. Prediction is performed by
linear discriminant function combining characteristics describing
donor and acceptor splice sites, 5'- and 3'-intron regions and also
coding region for each open reading frames flanked by GT and AG base
pairs.
Current version of the program predict only internal exons with GT and
AG conserved base pair for donor and acceptor splice sites, respectively.
They are usualy include more than 99% of the all authentic splice sites.
Further versions of the program will have option for other variants
and extention for other species.
Accuracy:
********************************
The accuracy of precise internal exon recognition on test set of 451 exons
is 77% with a specificity of 79%. The recognition quality computed at the
level of individual nucleotides is 88% for exons sequences with the
level 98% for intron sequences. This corresponds to a correlation coeffi-
cient of 0.87.
If you have a sequence containing a complete gene, 5'- or 3'-exons, you can
use gene structure prediction program FGENEH or exons recognition program
FGENEH from the server.
Submitting sequences via email:
***********************************
For email submission the sequences must have the following format:
Nane of your sequence
ccatctctgtcttgcaggacaatgccgtcttctgtctcgtggggcatcctcctgctggca
ggcctgtgctgcctggtccctgtctccctggctgaggatccccagggagatgctgcccag
aagacagatacatcccaccatgatcaggatcacccaaccttcaacaagatcacccccaac
ctggctgagttcgccttcagcctataccgccagctggcacaccagtccaacagcaccaat
atcttcttctccccagtgagcatcg...............
(Restrict the line length to 80 characters or less).
You could send the file containing the sequence to:
service at theory.bchs.uh.edu
Subject line must be:
hexon
Hexon output:
******************
1st line - name of your sequence
2nd line - length of your sequence
3d line - number of potential exons
4th line and next - positions and scores of predicted exons
It must be mentioned that some predicted exon may be partially
coinciding with real exons and the higher the score of an exon
the more probably it is a precise authentic exon.
For example:
HUMALPHA 4556 bp ds-DNA PRI 15-SEP-1
length of sequence - 4556
number of potential exon: 10
380 - 516 w= 9.66
611 - 727 w= 17.13
839 - 954 w= 21.43
1147 - 1321 w= 9.08
1819 - 1953 w= 12.21
2053 - 2125 w= 14.77
2254 - 2388 w= 12.91
2470 - 2661 w= 10.12
2881 - 2997 w= 17.17
3120 - 3562 w= 5.87
References:
The method is described in detail in:
Solovyev V.V.,Salamov A.A., Lawrence C.B.
Predicting internal exons by oligonucleotide composition and
discriminant analysis of spliceable open reading frames.
(Nucl.Acids Res.,1994, in press).
Solovyev V.V., Salamov A.A. , Lawrence C.B.
The prediction of human exons by oligonucleotide composition and
discriminant analysis of spliceable open reading frames.
in: The Second International conference on Intelligent systems
for Molecular Biology (eds. Altman R., Brutlag D.,
Karp R., Latrop R. and Searls D.), AAAI Press, Menlo Park, CA
(1994, in press)
Problems, comments, and suggestion:
can be mailed to solovyev at cmb.bcm.tmc.edu.
