Greengenes2 2022.10 Support

CHANGELOG.md

@@ -7,6 +7,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### `Added`
 
+- [#666](https://github.com/nf-core/ampliseq/pull/666) - Added Greengenes2 database, version 2022.10, support for QIIME2 taxonomic classification.
 - [#667](https://github.com/nf-core/ampliseq/pull/667) - Added `--qiime_ref_tax_custom` to permit custom reference database for QIIME2 taxonomic classification
 - [#674](https://github.com/nf-core/ampliseq/pull/674) - Add PhytoRef database for DADA2 taxonomy assignment using `--dada_ref_taxonomy phytoref`
 - [#675](https://github.com/nf-core/ampliseq/pull/675) - Add the Zehr lab nifH database for DADA2 taxonomy assignment using `--dada_ref_taxonomy zehr-nifh`

bin/taxref_reformat_qiime_greengenes2022.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+
+# Decompress files.
+gzip -c -d *.seqs.fna.gz > greengenes2.fna
+gzip -c -d *.taxonomy.md5.tsv.gz > greengenes2.tax

conf/base.config

@@ -63,4 +63,9 @@ process {
     withName:CUSTOM_DUMPSOFTWAREVERSIONS {
         cache = false
     }
+    withName:QIIME2_EXTRACT {
+        cpus   = { check_max( 12    * task.attempt, 'cpus'    ) }
+        memory = { check_max( 12.GB * task.attempt, 'memory'  ) }
+        time   = { check_max( 24.h  * task.attempt, 'time'    ) }
+    }
 }

conf/ref_databases.config

@@ -330,6 +330,18 @@ params {
             citation = "McDonald, D., Price, M., Goodrich, J. et al. An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J 6, 610–618 (2012). https://doi.org/10.1038/ismej.2011.139"
             fmtscript = "taxref_reformat_qiime_greengenes85.sh"
         }
+        'greengenes2' {
+            title = "Greengenes2 16S - Version 2022.10"
+            file = [ "http://ftp.microbio.me/greengenes_release/2022.10/2022.10.seqs.fna.gz", "http://ftp.microbio.me/greengenes_release/2022.10/2022.10.taxonomy.md5.tsv.gz" ]
+            citation = "McDonald, D., Jiang, Y., Balaban, M. et al. Greengenes2 unifies microbial data in a single reference tree. Nat Biotechnol (2023). https://doi.org/10.1038/s41587-023-01845-1"
+            fmtscript = "taxref_reformat_qiime_greengenes2022.sh"
+        }
+        'greengenes2=2022.10' {
+            title = "Greengenes2 16S - Version 2022.10"
+            file = [ "http://ftp.microbio.me/greengenes_release/2022.10/2022.10.seqs.fna.gz", "http://ftp.microbio.me/greengenes_release/2022.10/2022.10.taxonomy.md5.tsv.gz" ]
+            citation = "McDonald, D., Jiang, Y., Balaban, M. et al. Greengenes2 unifies microbial data in a single reference tree. Nat Biotechnol (2023). https://doi.org/10.1038/s41587-023-01845-1"
+            fmtscript = "taxref_reformat_qiime_greengenes2022.sh"
+        }
     }
     //Sintax taxonomic reference databases
     sintax_ref_databases {

docs/usage.md

@@ -225,6 +225,7 @@ Pre-configured reference taxonomy databases are:
 | sbdi-gtdb    | +     | -      | -       | -      | 16S rRNA                                      |
 | rdp          | +     | -      | +       | -      | 16S rRNA                                      |
 | greengenes   | -     | -      | +       | (+)²   | 16S rRNA                                      |
+| greengenes2  | -     | -      | -       | +      | 16S rRNA                                      |
 | pr2          | +     | -      | -       | -      | 18S rRNA                                      |
 | unite-fungi  | +     | +      | -       | +      | eukaryotic nuclear ribosomal ITS region       |
 | unite-alleuk | +     | +      | -       | +      | eukaryotic nuclear ribosomal ITS region       |

modules/local/qiime2_extract.nf

@@ -1,7 +1,5 @@
 process QIIME2_EXTRACT {
     tag "${meta.FW_primer}-${meta.RV_primer}"
-    label 'process_low'
-    label 'single_cpu'
 
     container "qiime2/core:2023.7"
 
@@ -20,6 +18,7 @@ process QIIME2_EXTRACT {
     if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
         error "QIIME2 does not support Conda. Please use Docker / Singularity / Podman instead."
     }
+    def args = task.ext.args ?: ''
     """
     export XDG_CONFIG_HOME="./xdgconfig"
     export MPLCONFIGDIR="./mplconfigdir"
@@ -37,9 +36,11 @@ process QIIME2_EXTRACT {
         --output-path ref-taxonomy.qza
     #Extract sequences based on primers
     qiime feature-classifier extract-reads \\
+        --p-n-jobs ${task.cpus} \\
         --i-sequences ref-seq.qza \\
         --p-f-primer ${meta.FW_primer} \\
         --p-r-primer ${meta.RV_primer} \\
+        $args \\
         --o-reads ${meta.FW_primer}-${meta.RV_primer}-ref-seq.qza \\
         --quiet
 

nextflow_schema.json

@@ -442,7 +442,9 @@
                         "unite-alleuk=8.3",
                         "unite-alleuk=8.2",
                         "unite-alleuk",
-                        "greengenes85"
+                        "greengenes85",
+                        "greengenes2",
+                        "greengenes2=2022.10"
                     ]
                 },
                 "qiime_ref_tax_custom": {