from https://www.whitelupin.fr/download.html
 wget https://www.whitelupin.fr/genome_files/Lalbus-20171117r1.genome.fasta
 wget https://www.whitelupin.fr/genome_files/mitochondrion/Lalb_mitochondrion.zip
 wget https://www.whitelupin.fr/genome_files/chloroplast/Lalb_chloroplast.zip
 wget https://www.whitelupin.fr/genome_files/Lalbus-20171117r1-v1.annot.gff3
 wget https://www.whitelupin.fr/genome_files/ShortStack_Clusters_All.gff3
 wget https://www.whitelupin.fr/genome_files/4-DANTE-tracks.gff
 wget https://www.whitelupin.fr/genome_files/Lalbus-satDNA-annotation-final.gff
 wget https://www.whitelupin.fr/genome_files/Lalbus-rDNA-annotation-final.gff
 wget https://www.whitelupin.fr/genome_files/gene.fasta
 wget https://www.whitelupin.fr/genome_files/prot.fasta
 wget https://www.whitelupin.fr/genome_files/cds.fasta
 wget https://www.whitelupin.fr/genome_files/mrna.fasta
 wget https://www.whitelupin.fr/genome_files/ncrna.fasta

 cat Lalbus-20171117r1.genome.fasta Lalb_chloroplast.fasta Lalb_mitochondrion.fasta > genome.fasta
 normalizer prefix --target genome.fasta --genus Lupinus --species albus --gnm 1 --infra_id Amiga --unique_key F4NR
 normalizer index --target Lupinus_albus/Amiga.gnm1.F4NR/lupal.Amiga.gnm1.F4NR.genome_main.fna

#mostly this is addressing various complaints of chado gff loader, and making best effort to
#get records linked to appropriate parents
 cat Lalb_chloroplast.gff | sed 's/NCBI Feature Key/ncbi_feature_key/;
 s/NCBI Join Type/ncbi_join_type/;
 s/Transferred /transferred_/g;
 s/\t\(gene\t.*\)pseudo=;/\tpseudo\1/;
 /pseudogene/ s/\<gene/pseudogene/g;
 s/Identical Bases/identical_Bases/;
 s/Mismatched Bases/mismatched_Bases/;
 s/Frequency/frequency/;
 s/Percent Identical/percent_Identical/;
 s/All Locations/all_Locations/;
 s/\t\([tr]\)rna\t/\t\1RNA\t/;
 s/\tcds\t/\tCDS\t/;
 s/ID=[^;]*//;
 s/Name=\([^ ;]*\) \([^ ;]\)/Name=\1_\2/;
 s/Name=\([^;]*\)/ID=\1;Name=lupal.\1/;
 /CDS/ s/ID=\([^;]*\)CDS/ID=\1CDS;Parent=\1gene/;
 /tRNA/ s/ID=\([^;]*\)tRNA/ID=\1tRNA;Parent=\1gene/;' | sort_gff.pl | perl -e '
 my %ids;
 while (<>) {
        my ($id) = /ID=([^;]*)/;
        my $count = $ids{$id};
        $count++;
        $ids{$id} = $count;
        $id .= "_$count";
        s/ID=[^;]*/ID=$id/;
        my ($parent) = /Parent=([^;]*)/;
        if (defined $parent) {
                my $count = $ids{$parent};
                $parent .= "_$count";
                s/Parent=[^;]*/Parent=$parent/;
        }
        print;
}
' | /erdos/adf/sw/hacks/gff_fixphase.pl | sort_gff.pl > chloro_for_gt

#similar to above, but much less custom munging needed here
cat Lalb_mitochondrion.gff3 | sed 's/ID=\([^;]*\)/ID=\1;Name=lupal.\1/; /CDS/ s/ID=cds\([^;]*\)/ID=cds\1;Parent=gene\1/;' | /erdos/adf/sw/hacks/gff_fixphase.pl | sort_gff.pl > mito_for_gt


#mostly addressing our naming convention additions, specifically things that are currently
#outside the scope of the ID prefixing done by the "normalizer"
cat Lalbus-20171117r1-v1.annot.gff3 | sed '
s/EC_number=/ec_number=/;
s/ID=mRNA:\([^;]*\)/ID=mRNA:\1.1/;
s/Parent=mRNA:\([^;]*\)/Parent=mRNA:\1.1/;
s/Name=\([^;]*\),\(Lalb[^;,]*\)/Name=\2;Alias=\1/;
s/Name=\([^;]*\)/Name=lupal.\1/;
/\tmRNA\t/ s/Name=\([^;]*\)/Name=\1.1/;
' > nuclear_for_gt

#putting it all together
cat nuclear_for_gt <(grep -vh '##gff-version' chloro_for_gt mito_for_gt) > combined.gff3
normalizer prefix --target  combined.gff3 --genus Lupinus --species albus --gnm 1 --ann 0 --infra_id Amiga --unique_key 3GKS

perl -p -e 's/>([^\t]*)/>lupal.Amiga.gnm1.ann1.mRNA:\1.1/' mrna.fasta > lupal.Amiga.gnm1.ann1.3GKS.mRNA.fna
perl -p -e 's/>([^\t]*)/>lupal.Amiga.gnm1.ann1.mRNA:\1.1/' prot.fasta > lupal.Amiga.gnm1.ann1.3GKS.protein.faa
perl -p -e 's/>([^\t]*)/>lupal.Amiga.gnm1.ann1.mRNA:\1.1/' cds.fasta > lupal.Amiga.gnm1.ann1.3GKS.cds.fna