source `$b/ROOT/bin/argos-env`
set em=/bio/bio-grid/mb/EVidenceModeler
set pa=/bio/bio-grid/mb/PASA/
set td=/bio/bio-grid/dpulex/prots/
set aug=/bio/bio-grid/mb/augustus
setenv EVM_HOME $em
setenv PASA_HOME $pa
set path= ( $path $PASA_HOME/bin )
setenv AUGUSTUS_CONFIG_PATH $aug/config/

#-----------
set rname=pasa_dmoj

set dgenome=dmoj_caf060210.fa
# set dgenome=dpulex_jgi060905.hardmask.fa
set dest1=dmoj-est.fa
#? need genes only if want to check/validate predictions
set dgenes=dmoj-genes.gff

touch log.$rname

echo "# Transcript alignments followed by alignment assembly"  \
>> log.$rname

../scripts/Launch_PASA_pipeline.pl -c alignAssembly.config \
-C -R --USE_GMAP \
-g $dgenome -t $dest1 \
>> & log.$rname

# not used: -T -u all_transcripts.fasta -f FL_accs.txt 

## optional
echo "# Importing the latest annotations into the PASA database"  \
>> log.$rname
## optional
../scripts/Load_Current_Gene_Annotations.dbi -c alignAssembly.config \
-g $dgenome  -P $dgenes \
>> & log.$rname

## optional
echo "# Performing an annotation comparison"  \
>> log.$rname

../scripts/Launch_PASA_pipeline.pl -c annotCompare.config \
-A -g $dgenome  -t $dest1 \
>> & log.$rname

perl -pi -e's/ID=chain/Parent=chain/;' $rname.pasa_assemblies.gff3

echo "# Updating our gene structure annotations"  \
>> log.$rname

../scripts/cDNA_annotation_updater.dbi \
-M "${rname}:localhost-port=3306-mysql_socket=/tmp/fbmysql.sock:gilbertd:darwin" -P null \
>> & log.$rname

#.... output training genes

$pa/scripts/pasa_asmbls_to_training_set.dbi \
-g $dgenome \
-M "${rname}:localhost-port=3306-mysql_socket=/tmp/fbmysql.sock" \
-p "gilbertd:darwin"   \
  >> log.train.$rname


#.....

echo "## Identification and Classification of All Alternative Splicing Variations" \
>> log.$rname

../scripts/Launch_PASA_pipeline.pl -c alignAssembly.config \
-g $dgenome -t $dest1  --ALT_SPLICE \
>> & log.$rname

#... if wanted; same as -A gff ...
echo "# dump current annots to gff" \
>> log.$rname

## this regurgitates the -A compare .gff output
../scripts/dump_valid_annot_updates.dbi \
-M "${rname}:localhost-port=3306-mysql_socket=/tmp/fbmysql.sock" \
-p "gilbertd:darwin"   \
>> & log.$rname


#......... annotCompare for FL Full Length to predict new genes ......

# ^^ tweak annotCompare with is_fli to do novel-gene annot ??
# cDNA_annotation_comparer.dbi
# dump_valid_annot_updates.dbi
#.......... 
#  test with scaffold99 (and some est asm not on genes: asm0gene/)

1. mysql.clusters drives annotCompare loop, replaced w/ temp = scaffold_99 subset

2. test as is

../scripts/cDNA_annotation_comparer.dbi -G dpulex_jgi060905a.fa \
 -M pasa_dapha:localhost-port=3306-mysql_socket=/tmp/fbmysql.sock \
 -p gilbertd:darwin \
 > output.annotation_compare.s99.out

3. set some cluster_link .is_fli = 1 and rerun
  on scaf99 without genes: asmbl_18445 asmbl_18490
  update cluster_link set is_fli = 1 where cdna_acc = 'asmbl_18445';

3b. set all to .is_fli = 1 where is_assembly = 1
3c.
../scripts/cDNA_annotation_comparer.dbi -G dpulex_jgi060905a.fa \
 -M pasa_dapha:localhost-port=3306-mysql_socket=/tmp/fbmysql.sock \
 -p gilbertd:darwin \
 > output.annotation_compare.allfli.out
 
 
4. output novel gene models :

../scripts/dump_valid_annot_updates.dbi  \
 -M pasa_dapha:localhost-port=3306-mysql_socket=/tmp/fbmysql.sock \
 -p gilbertd:darwin \
 > output.annotation_compare.s99fli.gff3


5. update is_fli = 0 for failed cdna tests

| cluster_id | cdna_acc | id     | is_assembly | is_fli | length | header | is_EST |

select cl.cdna_acc, cl.id, cl.is_assembly, cl.is_fli, s.status_id
from cluster_link cl, status_link s
where cl.is_assembly = 1 and 
(s.cdna_acc = cl.cdna_acc and s.status_id = 1)


update cluster_link cl, status_link s
set cl.is_fli = 0 
where cl.is_assembly = 1 and  
(s.cdna_acc = cl.cdna_acc and s.status_id = 1)

-- also this: short est not fli: status:23
update cluster_link cl, status_link s
set cl.is_fli = 0 
where cl.is_assembly = 1 and 
(s.cdna_acc = cl.cdna_acc and s.status_id = 23)