Skip to content

Commit

Permalink
fix #122. GTF to GFF create chunck of CDS for stop codon starting in …
Browse files Browse the repository at this point in the history 
…exon. (#123)
  • Loading branch information
@Juke34
Juke34 authored and Jacques Dainat committed Apr 27, 2021
1 parent 32d6aeb commit 6d6a220
Showing 1 changed file with 51 additions and 5 deletions.
56 changes: 51 additions & 5 deletions lib/AGAT/OmniscientI.pm
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1705,6 +1705,7 @@ sub _check_cds{
}
if($size_stop > 3){
dual_print($log, "Warning: $id_l2 has several stop_codon\n", $verbose);
next;
}
else{
$codon_split = 1;
Expand All @@ -1720,9 +1721,9 @@ sub _check_cds{

if($cds->end == $stop->end){next;} # Everything looks fine the stop codon is within the CDS

if($cds->end +1 != $stop->start){
if($cds->end +1 != $stop->start){ # Last stop in not adjacent to CDS.

if($codon_split){
if($codon_split){ # Not adjacent because stop splited

if ($cds->end + 1 == ($list_stop[0]->start) ){
dual_print($log, "Extend CDS to the first part of the stop codon\n", 0); #print log only
Expand All @@ -1739,8 +1740,31 @@ sub _check_cds{
push (@{$hash_omniscient->{"level3"}{'cds'}{$id_l2}}, $new_cds);
}
}
else{dual_print($log, "Warning: $id_l2 stop codon not adjacent to the CDS\n", $verbose);}
else{
# Not adjacent because stop start a new exon
my $stop_start_exon=undef;
if( exists_keys($hash_omniscient,('level3', 'exon', $id_l2)) ){
foreach my $l3_feature ( @{$hash_omniscient->{'level3'}{'exon'}{$id_l2}} ){
if ( $l3_feature->start == $stop->start ){
if ( $l3_feature->end >= $stop->end ){ # check stop if fully within a CDS
# create the cds chunk missing
my $new_cds = clone($cds);
$new_cds->start($stop->start);
$new_cds->end($stop->end);
my $size_stop = $stop->end - $stop->start + 1;
$new_cds->frame(3 - $size_stop);
my $uID = _create_ID($miscCount, $uniqID, $uniqIDtoType, 'cds', $new_cds->_tag_value('ID'), PREFIX_NEW_ID); #method will push the uID
create_or_replace_tag($new_cds, 'ID', $uID); # remove parent ID because, none.
push (@{$hash_omniscient->{"level3"}{'cds'}{$id_l2}}, $new_cds);
$stop_start_exon=1;
}
}
}
}
dual_print($log, "Warning: $id_l2 stop codon not adjacent to the CDS\n", $verbose) if (! $stop_start_exon);
}
}
# stop is adjacent
else{
$cds->end($stop->end);
$resume_case++;
Expand Down Expand Up @@ -1769,7 +1793,29 @@ sub _check_cds{
push (@{$hash_omniscient->{"level3"}{'cds'}{$id_l2}}, $new_cds);
}
}
else{ dual_print($log, "Warning: $id_l2 stop codon not adjacent to the CDS\n", $verbose);}
else{
# Not adjacent because stop start a new exon
my $stop_start_exon=undef;
if( exists_keys($hash_omniscient,('level3', 'exon', $id_l2)) ){
foreach my $l3_feature (@{$hash_omniscient->{'level3'}{'exon'}{$id_l2}}){
if ( $l3_feature->end == $stop->end ){
if ( $l3_feature->start <= $stop->start ){ # check stop if fully within a CDS
# create the cds chunk missing
my $new_cds = clone($cds);
$new_cds->start($stop->start);
$new_cds->end($stop->end);
my $size_stop = $stop->end - $stop->start + 1;
$new_cds->frame(3 - $size_stop);
my $uID = _create_ID($miscCount, $uniqID, $uniqIDtoType, 'cds', $new_cds->_tag_value('ID'), PREFIX_NEW_ID); #method will push the uID
create_or_replace_tag($new_cds, 'ID', $uID); # remove parent ID because, none.
push (@{$hash_omniscient->{"level3"}{'cds'}{$id_l2}}, $new_cds);
$stop_start_exon=1;
}
}
}
}
dual_print($log, "Warning: $id_l2 stop codon not adjacent to the CDS\n", $verbose) if (! $stop_start_exon);
}
}
else{
$cds->start($stop->start);
Expand Down Expand Up @@ -3194,7 +3240,7 @@ sub select_gff_format{

my $Ninethcolum = $1;

#replace value quoted by a string in case some special character are within the quote e.g. = ;
#replace value quoted by a string in case some special character are within the quote e.g. = ;
$_ = $Ninethcolum;
s/("([^"]|"")*")/quoted_value/g;
$Ninethcolum = $_;
Expand Down

0 comments on commit 6d6a220

Please sign in to comment.