USING MAIZE GENE FINDER
The parameters used in this program are based on the author's analysis of 15
Z. mays genes containing 74 introns. The numerical values assigned to the
various parameters were determined empirically. Continuing studies of gene sequences
may result in further optimization of these values.
Intron prediction is based primarily on the following factors:
1) The context of the 5' intron junction (...GT...).
2) The context of the 3' junction (...AG...).
3) The context of the Branch Point.
4) The distance between the Branch Point and The 3' junction.
5) The length of the resultant intron.
6) The nucleotide compositions of intron and exon adjacent to the 3' junction.
As of July 2000 this program is somewhat less successful at predicting introns in
maize genomic sequences than its parent program is at predicting introns in
Aspergillus niger sequences. This results in longer processing time and the
prediction of more false Open Reading Frames (ORFs). Since processing time is
limited by my web server, this program may not complete processing your sequence;
especially if it is very long and/or rich in potential ORFs. Therefore the user
is given several options which might circumvent this problem.
USER-SELECTED PARAMETERS
ATG selection
It is often possible to predict which ATGs are likely to be used by valid ORFs.
The user may choose to process only the most-likely ATGs, thereby reducing both
processing time and invalid ORFs. This option is permitted only for ATGs downstream
of nt 1000 (as valid ATG prediction depends on upstream sequences.) Thus, a second
submission would be required to analyze all the ATGs in the 1000 nt at the 5' end.
(The ATG position of 8 of the maize genes used in this study was > 1000. These
ATGs were rated: a, b, c, d, e, f, g, and h).
Processing Mode
By default, the program will attempt to identify the best ORFs generated from each
ATG (as specified by the user). If the submitted sequence is long and/or rich in
potential ORFs, the server may kill the job before it is done (and the user gets
no information.) If this happens, the user may choose to FIRST identify the ATGs
from which ORFs longer than a specified cutoff can be generated. (Since the program
stops processing an ATG once an ORF > cutoff is found, the larger cutoff is, the
longer the program will run.) The user may then analyze each ATG individually.
OUTPUT
last modified 6/7/00
Comments or suggestions:
bawill@molecularworkshop.com
