New script to move feature from features of a same record (agat_sp_mo…

…ve_attributes_within_records) (#413)

* New script to move attributes from features of a same record, add test and doc

* fix create_or_append_tag in Omniscient Tool to take more than the first value when the attribute has several values

---------

Co-authored-by: Jacques Dainat <[email protected]>

docs/index.rst

@@ -103,6 +103,7 @@ Contents
    tools/agat_sp_manage_functional_annotation.md
    tools/agat_sp_manage_introns.md
    tools/agat_sp_merge_annotations.md
+   tools/agat_sp_move_attributes_within_records
    tools/agat_sp_prokka_fix_fragmented_gene_annotations.md
    tools/agat_sp_sensitivity_specificity.md
    tools/agat_sp_separate_by_record_type.md

docs/tools/agat_sp_move_attributes_within_records.md

@@ -0,0 +1,67 @@
+# NAME
+
+agat\_sp\_move\_attributes\_within\_records.pl
+
+# DESCRIPTION
+
+The script aims to keep move attributes within a record e.g. from Level1 to Level2 and/or Level3 features; and / or from Level2 to Level2 or Level3 features; and / or from Level3 to Level3 features.
+Example of L1 feature: gene
+Example of L2 featrue
+
+# SYNOPSIS
+
+```
+agat_sp_move_attributes_within_records.pl --gff infile.gff --feature_copy mRNA  --feature_paste CDS --attribute Dbxref,Ontology [ --output outfile ]
+agat_sp_move_attributes_within_records.pl --help
+```
+
+# OPTIONS
+
+- **-f**, **--reffile**, **--gff**  or **-ref**
+
+    Input GFF3 file that will be read
+
+- **--feature\_copy** or **--fc**
+
+    primary tag (feature type) option to list from which feature we will copy the attributes, case insensitive. 
+    You can specified a feature (or a coma separated list) by giving its primary tag / feature type (column 3) value as: cds, Gene, MrNa, etc
+    You can specify directly all the feature of a particular level:
+          level2=mRNA,ncRNA,tRNA,etc
+          level3=CDS,exon,UTR,etc
+    By default all level2 feature are used. 
+
+- **--feature\_paste** or **--fp**
+
+    primary tag (feature type) option to list to which feature we will paste the attributes, case sensitive. 
+    You can specified a feature (or a coma separated list) by giving its primary tag / feature type (column 3) value as: cds, Gene, MrNa, etc
+    You can specify directly all the feature of a particular level:
+          level2=mRNA,ncRNA,tRNA,etc
+          level3=CDS,exon,UTR,etc
+    By default all feature level3 are used. 
+
+- **-a** or **--attribute**
+
+    Attribute that will be copied and pasted. Case sensitive.
+    You can specified an attribute (or a coma separated list) by giving its attribute tag value (column9) as: Ontology, Dbxref, etc
+    Default: all\_attributes
+    /!\\ <all\_attributes> is a specific parameter meaning all the attributes will be use.
+
+- **-o** or **--output**
+
+    Output GFF file.  If no output file is specified, the output will be
+    written to STDOUT.
+
+- **-v**
+
+    Verbose option for debugging purpose.
+
+- **-c** or **--config**
+
+    String - Input agat config file. By default AGAT takes as input agat\_config.yaml file from the working directory if any, 
+    otherwise it takes the orignal agat\_config.yaml shipped with AGAT. To get the agat\_config.yaml locally type: "agat config --expose".
+    The --config option gives you the possibility to use your own AGAT config file (located elsewhere or named differently).
+
+- **-h** or **--help**
+
+    Display this helpful text.
+

lib/AGAT/OmniscientTool.pm

@@ -571,7 +571,8 @@ sub merge_overlap_loci{
 							my @list_tag_l2 = $omniscient->{'level1'}{$tag_l1}{$id2_l1}->get_all_tags();
 							foreach my $tag (@list_tag_l2){
 								if(lc($tag) ne "parent" and lc($tag) ne "id"){
-									create_or_append_tag($omniscient->{'level1'}{$tag_l1}{$id_l1}, $tag ,$omniscient->{'level1'}{$tag_l1}{$id2_l1}->get_tag_values($tag));
+									my @tag_values = $omniscient->{'level1'}{$tag_l1}{$id2_l1}->get_tag_values($tag);
+									create_or_append_tag($omniscient->{'level1'}{$tag_l1}{$id_l1}, $tag , \@tag_values);
 								}
 							}
 							# remove the level1 of the ovelaping one
@@ -611,7 +612,8 @@ sub merge_overlap_loci{
                       						$resume_identic++;
 											my @list_tag_l2 = $common->get_all_tags();
 											foreach my $tag (@list_tag_l2){
-												create_or_append_tag($kept_l2, "merged_".$tag ,$common->get_tag_values($tag));
+												my @tag_values = $common->get_tag_values($tag);
+												create_or_append_tag($kept_l2, "merged_".$tag , \@tag_values);
 											}
 										}
 									}
@@ -1325,6 +1327,9 @@ sub create_or_replace_tag{
 
 # INPUT: feature object, String tag, String or Array ref;
 # Output: None
+# /!\ If values are extracted using get_tag_values($tag) you should first save the result in an array and send the array ref to this function e.g
+# 	my @tag_values = $feature->get_tag_values($tag);
+#	create_or_append_tag($other_feature, $tag , \@tag_values);
 sub create_or_append_tag{
 	my ($feature, $tag, $value)=@_;
 
@@ -1333,15 +1338,15 @@ sub create_or_append_tag{
 				my @original_values = $feature->get_tag_values($tag);
 				foreach my $value (@{$value}){
 					if(! grep { $value eq $_ } @original_values){
-						$feature->add_tag_value($tag,@{$value});
+						$feature->add_tag_value($tag,$value);
 					}
 				}
 			}
 			else{
-						my @original_values = $feature->get_tag_values($tag);
+				my @original_values = $feature->get_tag_values($tag);
 	        	if(! grep { $value eq $_ } @original_values){
 	        		$feature->add_tag_value($tag,$value);
-  					}
+  				}
         	}
 	}
 	else{

t/gff_syntax/out/15_correct_output.gff

@@ -9,4 +9,4 @@ scaffold789	maker	match_part	558184	560123	.	+	.	ID=agat-exon-4;Parent=CLUHART00
 scaffold789	maker	match_part	561401	561519	.	+	.	ID=agat-exon-5;Parent=CLUHART00000006146;Target=CLUHART00000006146 1941 2059;merged_ID=CLUHART00000006146:exon:996;merged_Parent=CLUHART00000006147;merged_Target=CLUHART00000006147 1941 2059
 scaffold789	maker	match_part	562057	562121	.	+	.	ID=CLUHART00000006147:exon:999;Parent=CLUHART00000006146;Target=CLUHART00000006147 2060 2124
 scaffold789	maker	match_part	564171	564235	.	+	.	ID=CLUHART00000006146:exon:997;Parent=CLUHART00000006146;Target=CLUHART00000006146 2060 2124
-scaffold789	maker	match_part	564372	564780	.	+	.	ID=agat-exon-6;Parent=CLUHART00000006146;Target=CLUHART00000006146 2125 2533;merged_ID=CLUHART00000006146:exon:998;merged_Parent=CLUHART00000006147;merged_Target=CLUHART00000006147
+scaffold789	maker	match_part	564372	564780	.	+	.	ID=agat-exon-6;Parent=CLUHART00000006146;Target=CLUHART00000006146 2125 2533;merged_ID=CLUHART00000006146:exon:998;merged_Parent=CLUHART00000006147;merged_Target=CLUHART00000006147,2125,2533

t/scripts_output.t

@@ -603,6 +603,15 @@ system(" $script --gff $input_folder/agat_sp_merge_annotations/fileA.gff  --gff
 ok( system("diff $result $outtmp") == 0, "output $script");
 unlink $outtmp;
 
+# ------------------- check agat_sp_move_attributes_within_records script-------------------
+
+$script = $script_prefix."bin/agat_sp_move_attributes_within_records.pl";
+$result = "$output_folder/agat_sp_move_attributes_within_records.gff";
+system(" $script --gff $input_folder/agat_sp_move_attributes_within_records.gff --fp exon,CDS --fc mRNA -o $outtmp 2>&1 1>/dev/null");
+#run test
+ok( system("diff $result $outtmp") == 0, "output $script");
+unlink $outtmp;
+
 # ------------------- check agat_sp_prokka_fragmented_gene_annotations script-------------------
 
 $script = $script_prefix."bin/agat_sp_prokka_fix_fragmented_gene_annotations.pl";

t/scripts_output/in/agat_sp_move_attributes_within_records.gff

@@ -0,0 +1,3 @@
+ptg000002l	AUGUSTUS	mRNA	3255	4626	0.5	+	.	ID=NBISM00000000001;Parent=NBISG00000000001;Dbxref=CDD:cd07067,Gene3D:G3DSA:3.40.50.1240,InterPro:IPR013078,InterPro:IPR029033,;Name=ARB_03491;Ontology_term=-;makerName=g1.t1;product=Probable phosphoglycerate mutase ARB_03491;uniprot_id=D4B4V1
+ptg000002l	AUGUSTUS	CDS	3255	3275	0.98	+	0	ID=NBISC00000000001;Parent=NBISM00000000001;makerName=g1.t1.CDS1
+

t/scripts_output/out/agat_sp_move_attributes_within_records.gff

@@ -0,0 +1,6 @@
+##gff-version 3
+ptg000002l	AGAT	gene	3255	4626	.	+	.	ID=NBISG00000000001;Dbxref=CDD:cd07067,Gene3D:G3DSA:3.40.50.1240,InterPro:IPR013078,InterPro:IPR029033;Name=ARB_03491;Ontology_term=-;makerName=g1.t1;product=Probable phosphoglycerate mutase ARB_03491;uniprot_id=D4B4V1
+ptg000002l	AUGUSTUS	mRNA	3255	4626	0.5	+	.	ID=NBISM00000000001;Parent=NBISG00000000001;Dbxref=CDD:cd07067,Gene3D:G3DSA:3.40.50.1240,InterPro:IPR013078,InterPro:IPR029033;Name=ARB_03491;Ontology_term=-;makerName=g1.t1;product=Probable phosphoglycerate mutase ARB_03491;uniprot_id=D4B4V1
+ptg000002l	AGAT	exon	3255	4626	.	+	.	ID=agat-exon-1;Parent=NBISM00000000001;Dbxref=CDD:cd07067,Gene3D:G3DSA:3.40.50.1240,InterPro:IPR013078,InterPro:IPR029033;Name=ARB_03491;Ontology_term=-;makerName=g1.t1.CDS1,g1.t1;product=Probable phosphoglycerate mutase ARB_03491;uniprot_id=D4B4V1
+ptg000002l	AUGUSTUS	CDS	3255	3275	0.98	+	0	ID=NBISC00000000001;Parent=NBISM00000000001;Dbxref=CDD:cd07067,Gene3D:G3DSA:3.40.50.1240,InterPro:IPR013078,InterPro:IPR029033;Name=ARB_03491;Ontology_term=-;makerName=g1.t1.CDS1,g1.t1;product=Probable phosphoglycerate mutase ARB_03491;uniprot_id=D4B4V1
+ptg000002l	AGAT	three_prime_UTR	3276	4626	.	+	.	ID=agat-three_prime_utr-1;Parent=NBISM00000000001;makerName=g1.t1.CDS1