926df1df02bdcf9a2b3d567a7b3483015431db6d
[qpalma.git] / scripts / est2gff_pipeline.sh
1 #!/bin/bash
2
3 #
4 # This script takes the alignment output of QPalma and uses the script psl2gff
5 # to cluster the reads together in order to infer gene structures.
6 #
7 # The result is a (are several) gff file(s) that are used as input to the
8 # compute_splicegraph script of the genebuild project.
9 #
10
11 g_config=/fml/ag-raetsch/share/databases/genomes/A_thaliana/arabidopsis_tair7/genebuild/genome.config
12
13 # First we count the numbers of incorrect gt and gc positions for the full
14 # alignment set not filtered by coverage numbers etc.
15
16 alignment_file=/fml/ag-raetsch/home/fabio/tmp/transcriptome_data/full.align.consistent.unique
17
18 for CHR in "CHR1" "CHR2" "CHR3" "CHR4" "CHR5"
19 do
20 for STRAND in "+" "-"
21 do
22 echo $CHR $STRAND
23 result1_fn=/fml/ag-raetsch/home/fabio/tmp/transcriptome_data/SpliceGraphResults/gff/test_myexons_${CHR}${STRAND}.gff
24 result2_fn=/dev/null
25
26 /fml/ag-raetsch/home/fabio/svn/projects/splicing/gff/bin/psl2gff EST $alignment_file $CHR ${STRAND} $result1_fn $result2_fn $g_config 1>>LOG
27 done
28 done