More proper build

.github/workflows/unit-testing.Listeria.Kraken1.yml

@@ -1,7 +1,7 @@
 # This is a subsampling unit test to get early results
 on: 
   push:
-    branches: [master]
+    branches: [master, dev, build-taxonomy]
 name: Listeria-with-Kraken1
 
 env:
@@ -23,7 +23,7 @@ jobs:
           perl-version: ${{ matrix.perl }}
           multi-thread: "true"
       - name: checkout my repo
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
         with:
           path: Kalamari
 
@@ -48,7 +48,9 @@ jobs:
           perl -MNet::FTP -e '$ftp = new Net::FTP("ftp.ncbi.nlm.nih.gov", Passive => 1); $ftp->login; $ftp->binary; $ftp->get("/entrez/entrezdirect/edirect.tar.gz");'
           gunzip -cv edirect.tar.gz | tar xf -
           rm -v edirect.tar.gz
-          export PATH=${PATH}:$HOME/edirect >& /dev/null || setenv PATH "${PATH}:$HOME/edirect"
+          echo $GITHUB_WORKSPACE/edirect >> $GITHUB_PATH
+          echo $GITHUB_WORKSPACE/Kalamari/bin >> $GITHUB_PATH
+          #export PATH=${PATH}:$HOME/edirect >& /dev/null || setenv PATH "${PATH}:$HOME/edirect"
           yes Y | ./edirect/setup.sh
           tree edirect
       - name: check-env
@@ -64,10 +66,6 @@ jobs:
         run:  perl Kalamari/bin/downloadKalamari.pl --outdir ${{ env.OUTDIR }} ${{ env.TSV }}
       - name: check-results
         run:  tree ${{ env.OUTDIR }}
-      #- name: download-more
-      #  run:  perl Kalamari/bin/downloadKalamari.pl --outdir ${{ env.OUTDIR }} ${{ env.TSV }} --and protein --and nucleotide
-      #- name: check-results
-      #  run:  tree ${{ env.OUTDIR }}
       - name: install kraken
         run:  |
           wget https://github.com/DerrickWood/kraken/archive/refs/tags/v1.1.1.tar.gz -O kraken-v1.1.1.tar.gz
@@ -76,12 +74,17 @@ jobs:
           chmod -v +x kraken-1.1.1/kraken-src/*
           echo $(realpath kraken-1.1.1/kraken-src) >> $GITHUB_PATH
           tree $(realpath) kraken-1.1.1
+      - name: build taxonomy
+        run: |
+          export PATH=$PATH:Kalamari/bin
+          buildTaxonomy.sh
+          ls -lh Kalamari/share
       - name: Kraken1 database
         run:  |
           echo $PATH
           which kraken-build
           mkdir -pv kraken
-          cp -rv Kalamari/src/taxonomy kraken/taxonomy
+          cp -rv Kalamari/share/kalamari-*/taxonomy kraken/taxonomy
           find ${{ env.OUTDIR }} -name '*.fasta' -exec kraken-build --db kraken --add-to-library {} \;
           tree kraken
           # Some super debugging here with -x

.github/workflows/unit-testing.Yersinia.Kraken2.yml

@@ -1,14 +1,12 @@
 # This is a subsampling unit test to get early results
 on: 
   push:
-    branches: [master]
-name: Yersinia-with-Kraken2
+    branches: [master, dev, build-taxonomy]
+name: Genera-with-Kraken2
 
 env:
   TSV: "Kalamari/src/genus.tsv"
-  OUTDIR: "Yersinia.out"
   DB: "kraken2"
-  SRC_TAX: "Kalamari/src/taxonomy"
   SRC_CHR: "Kalamari/src/chromosomes.tsv"
   SRC_PLD: "Kalamari/src/plasmids.tsv"
   GENUS: Yersinia
@@ -20,15 +18,16 @@ jobs:
       matrix:
         os: ['ubuntu-20.04' ]
         perl: [ '5.32' ]
-    name: Perl ${{ matrix.perl }} on ${{ matrix.os }}
+        GENUS: [ 'Yersinia', 'Listeria']
+    name: ${{ matrix.GENUS }} Perl ${{ matrix.perl }} on ${{ matrix.os }}
     steps:
       - name: Set up perl
         uses: shogo82148/actions-setup-perl@v1
         with:
           perl-version: ${{ matrix.perl }}
           multi-thread: "true"
       - name: checkout my repo
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
         with:
           path: Kalamari
 
@@ -40,29 +39,37 @@ jobs:
       - name: select for only for this genus
         run:  |
           head -n 1 ${{ env.SRC_CHR }} > ${{ env.TSV }}
-          grep -m 2 ${{ env.GENUS }} ${{ env.SRC_CHR }} >> ${{ env.TSV }}
-          grep -m 2 ${{ env.GENUS }} ${{ env.SRC_PLD }} >> ${{ env.TSV }}
-          echo "These are the ${{ env.GENUS }} genomes for downstream tests"
+          grep -m 2 ${{ matrix.GENUS }} ${{ env.SRC_CHR }} >> ${{ env.TSV }}
+          grep -m 2 ${{ matrix.GENUS }} ${{ env.SRC_PLD }} >> ${{ env.TSV }}
+          echo "These are the ${{ matrix.GENUS }} genomes for downstream tests"
           column -ts $'\t' ${{ env.TSV }}
           hexdump -c ${{ env.TSV }}
       - name: download
-        run:  perl Kalamari/bin/downloadKalamari.pl --outdir ${{ env.OUTDIR }} ${{ env.TSV }}
+        run:  perl Kalamari/bin/downloadKalamari.pl --outdir ${{ matrix.GENUS }} ${{ env.TSV }}
       - name: check-results
-        run:  tree ${{ env.OUTDIR }}
+        run:  |
+          tree ${{ matrix.GENUS }}
+          echo "First two lines of each fasta file:"
+          find ${{ matrix.GENUS }} -name '*.fasta' | xargs head -n 2 | cut -c 1-60
       - name: install kraken
         run:  |
           wget https://github.com/DerrickWood/kraken2/archive/refs/tags/v2.1.2.tar.gz -O kraken-v2.1.2.tar.gz
           tar zxvf kraken-v2.1.2.tar.gz
           cd kraken2-2.1.2 && bash install_kraken2.sh target && cd -
           ls -lhS kraken2-2.1.2/target
           chmod +x kraken2-2.1.2/target/*
+      - name: build taxonomy
+        run: |
+          export PATH=$PATH:Kalamari/bin
+          buildTaxonomy.sh
+          ls -lh Kalamari/share
       - name: Kraken2 database
         run:  |
           export PATH=$PATH:kraken2-2.1.2/target
           which kraken2-build
           mkdir -pv ${{ env.DB }}
-          cp -rv ${{ env.SRC_TAX }} ${{ env.DB }}/taxonomy
-          find ${{ env.OUTDIR }} -name '*.fasta' -exec kraken2-build --db ${{ env.DB }} --add-to-library {} \;
+          cp -rv Kalamari/share/kalamari-*/taxonomy ${{ env.DB }}/taxonomy
+          find ${{ matrix.GENUS }} -name '*.fasta' -exec kraken2-build --db ${{ env.DB }} --add-to-library {} \;
           tree ${{ env.DB }}
           echo ".....Building the database....."
           kraken2-build --build --db ${{ env.DB }} --threads 2
@@ -71,10 +78,9 @@ jobs:
           export PATH=$PATH:kraken2-2.1.2/target
           tree ${{ env.DB }}
           ls -lhSR ${{ env.DB }}
-          QUERY=$(find ${{ env.OUTDIR }} -name '*.fasta' | head -n 1)
+          QUERY=$(find ${{ matrix.GENUS }} -name '*.fasta' | head -n 1)
           echo "QUERY is $QUERY"
-          head -n 2 $QUERY
-          kraken2 --db ${{ env.DB }} --report kraken2.report --use-mpa-style --output kraken2.raw $QUERY
+          set -x; kraken2 --db ${{ env.DB }} --report kraken2.report --use-mpa-style --output kraken2.raw $QUERY; set +x;
           head kraken2.report kraken2.raw
 
 

.github/workflows/unit-testing.yml

@@ -1,6 +1,6 @@
 on: 
   push:
-    branches: [master]
+    branches: [master, dev]
 name: Pull-down-all-accessions
 
 jobs:

.github/workflows/validateTaxonomy.yml

@@ -1,6 +1,6 @@
 on: 
   push:
-    branches: [fix-CI, master]
+    branches: [master, dev, build-taxonomy]
 name: Validate taxonomy
 
 jobs:
@@ -18,15 +18,26 @@ jobs:
           perl-version: ${{ matrix.perl }}
           multi-thread: "true"
       - name: checkout my repo
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
         with:
           path: Kalamari
-
-      - name: validate taxonomy
+      - name: update PATH
+        run:  |
+          echo $GITHUB_WORKSPACE/Kalamari/bin >> $GITHUB_PATH
+          echo $PATH
+          echo ""
+          cat $GITHUB_PATH
+      - name: build taxonomy
         run:  |
-          perl Kalamari/bin/validateTaxonomy.pl Kalamari/src
+          echo $PATH
+          bash Kalamari/bin/buildTaxonomy.sh
+          ls -lhR Kalamari/share/kalamari-*/taxonomy
+      #- name: validate taxonomy
+      #  run:  |
+      #    perl Kalamari/bin/validateTaxonomy.pl Kalamari/share/kalamari-*/taxonomy/nodes.dmp Kalamari/share/kalamari-*/taxonomy/names.dmp
       - name: matching taxids
         run:  |
+          export taxdir=$(\ls -d Kalamari/share/kalamari-*/taxonomy)
           echo "Making sure that all taxids in chromosomes.tsv and plasmids.tsv are present in nodes.tsv and names.tsv"
-          tail -n +2 Kalamari/src/chromosomes.tsv Kalamari/src/plasmids.tsv -q | perl -F'\t' -lane 'BEGIN{@node=`cat Kalamari/src/taxonomy/nodes.dmp`; for $n(@node){($taxid)=split(/\t/, $n); $taxid{$taxid}++; } } for my $t($F[2], $F[3]){ if(!$taxid{$t}){ print "Could not find $t taxid";} }'
-          tail -n +2 Kalamari/src/chromosomes.tsv Kalamari/src/plasmids.tsv -q | perl -F'\t' -lane 'BEGIN{@name=`cat Kalamari/src/taxonomy/names.dmp`; for $n(@name){($taxid)=split(/\t/, $n); $taxid{$taxid}++; } } for my $t($F[2], $F[3]){ if(!$taxid{$t}){ print "Could not find $t taxid";} }'
+          tail -n +2 Kalamari/src/chromosomes.tsv Kalamari/src/plasmids.tsv -q | perl -F'\t' -lane 'BEGIN{@node=`cat $ENV{taxdir}/nodes.dmp`; for $n(@node){($taxid)=split(/\t/, $n); $taxid{$taxid}++; } } for my $t($F[2], $F[3]){ if(!$taxid{$t}){ print "Could not find $t taxid";} }'
+          tail -n +2 Kalamari/src/chromosomes.tsv Kalamari/src/plasmids.tsv -q | perl -F'\t' -lane 'BEGIN{@name=`cat $ENV{taxdir}/names.dmp`; for $n(@name){($taxid)=split(/\t/, $n); $taxid{$taxid}++; } } for my $t($F[2], $F[3]){ if(!$taxid{$t}){ print "Could not find $t taxid";} }'

.gitignore

@@ -0,0 +1,2 @@
+edirect
+share

README.md

@@ -39,20 +39,22 @@ using your own email address instead of `[email protected]`.
 
 ## Download instructions
 
-For usage, run `perl bin/downloadKalamari.pl --help`
+First, build the taxonomy.
+The script `buildTaxonomy.sh` uses the diffs in Kalamari to enhance the default NCBI taxonomy.
+Next, `filterTaxonomy.sh` reduces the taxonomy files to just those found in Kalamari.
+`filterTaxonomy.sh` uses `taxonkit` and so this needs to be in your
+environment before starting.
 
-    SRC=Kalamari
-    perl bin/downloadKalamari.pl -o $SRC src/chromosomes.tsv
+    bash bin/buildTaxonomy.sh
+    bash bin/filterTaxonomy.sh
 
-### ...with plasmids
+To download the chromosomes and plasmids, use the `.tsv` files, respectively, with `downloadKalamari.pl`.
+Run `downloadKalamari.pl --help` for usage.
+However, to download the files to a standard location,
+please simply use `downloadKalamari.sh` which uses
+`downloadKalamari.pl` internally.
 
-    SRC=Kalamari
-    perl bin/downloadKalamari.pl -o $SRC src/chromosomes.tsv src/plasmids.tsv
-
-### taxonomy
-
-The taxonomy files `nodes.dmp` and `names.dmp` are under `src/taxonomy-VER` 
-where `VER` is the version of Kalamari.
+    bash bin/downloadKalamari.pl
 
 ## Database formatting instructions
 
@@ -80,4 +82,4 @@ Please see [CONTRIBUTING.md](CONTRIBUTING.md)
 
 ## Citation
 
-Please refer to the ASM 2018 poster under docs
+Please refer to the ASM 2018 poster under docs.

bin/buildKraken1.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+
+set -eu
+
+thisdir=$(dirname $0)
+KALAMARI_VER=$(downloadKalamari.pl --version)
+
+sharedir=$thisdir/../share/kalamari-$KALAMARI_VER
+SRC="$sharedir/kalamari"
+TAXDIR="$sharedir/taxonomy/filtered"
+
+# Test prereqs
+which kraken-build
+which jellyfish
+
+DB="$sharedir/kalamari-kraken1"
+mkdir -pv $DB
+cp -rv $TAXDIR $DB/taxonomy
+find $SRC -name '*.fasta' \
+  -exec kraken-build --db $DB --add-to-library {} \;
+kraken-build --db $DB --build --threads 1
+kraken-build --db $DB --clean

bin/buildKraken2.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+
+set -eu
+
+thisdir=$(dirname $0)
+KALAMARI_VER=$(downloadKalamari.pl --version)
+
+sharedir=$thisdir/../share/kalamari-$KALAMARI_VER
+SRC="$sharedir/kalamari"
+TAXDIR="$sharedir/taxonomy/filtered"
+
+# Test prereqs
+which kraken2-build
+which jellyfish
+
+DB="$sharedir/kalamari-kraken2"
+mkdir -pv $DB
+cp -rv $TAXDIR $DB/taxonomy
+find $SRC -name '*.fasta' \
+  -exec kraken2-build --db $DB --add-to-library {} \;
+kraken2-build --db $DB --build --threads 1
+kraken2-build --db $DB --clean