Merge pull request #112 from mhoban/zotu_identity

Add `--zotu-identity` parameter for zOTU table generation step

README.md

@@ -852,6 +852,7 @@ These options control how (and by what tool) sequences are denoised and zOTUs ar
 <small>**`--denoiser [tool/path]`**</small>:  Sets the tool used for denoising & chimera removal. Accepted options: 'usearch', 'usearch32', 'vsearch', path to 64-bit usearch executable (default: vsearch)    
 <small>**`--min-abundance [num]`**</small>:  Minimum sequence abundance for zOTU determination; sequences with abundances below the specified threshold will be discarded during the denoising process (default: 8)   
 <small>**`--alpha [num]`**</small>: Alpha parameter passed to the UNOISE3 algorithm (see the [unoise2 paper for more info](https://doi.org/10.1101/081257)) (default: 2.0)  
+<small>**`--zotu-identity [num]`**</small>: Fractional pairwise identity used to match raw reads to zOTUs, equivalent to `vsearch` `--id`/`usearch` `-id` parameters (default: 0.97)  
 <small>**`--usearch`**</small>:  Alias for `--denoiser usearch`  
 
 ### zOTU curation using LULU

lib/helper.groovy

@@ -253,6 +253,7 @@ class helper {
       --alpha [num]                 Sets the alpha parameter for the UNOISE3 algorithm (default: ${params.alpha})
       --min-abundance [num]         Minimum sequence abundance for zOTU determination; sequences below threshold will be discarded
                                     (default: ${params.minAbundance}) 
+      --zotu-identity [num]         Fractional identity (0–1) for zOTU search (default: 0.97)
       --usearch                     shortcut for --denoiser usearch
 
     LULU zOTU curation:

nextflow.config

@@ -65,6 +65,7 @@ params.usearch = false
 params.denoiser = params.usearch ? 'usearch' : 'vsearch'  
 params.minAbundance = 8
 params.alpha = 2.0
+params.zotuIdentity = 0.97
 try {
   params.execDenoiser = new File(params.denoiser).canExecute()
 } catch (Exception e) {

rainbow_bridge.nf

@@ -411,6 +411,7 @@ process dereplicate {
   output:
     tuple val(id), path("${id}_unique.fasta"), path("${id}_zotus.fasta"), path("zotu_table.tsv"), emit: result
     path 'settings.txt'
+    path 'zotu_map.tsv'
 
   script:
   if (params.denoiser == "vsearch") {
@@ -423,10 +424,24 @@ process dereplicate {
     # 3. get rid of chimeras
     # 4. match original sequences to zotus by 97% identity
     if [ -s "${relabeled_merged}" ]; then 
-      vsearch --threads ${task.cpus} --fastq_qmax ${params.maxQuality} --derep_fulllength ${relabeled_merged} --sizeout --output "${id}_unique.fasta"
-      vsearch --threads ${task.cpus} --fastq_qmax ${params.maxQuality} --cluster_unoise "${id}_unique.fasta" --centroids "${id}_centroids.fasta" --minsize ${params.minAbundance} --unoise_alpha ${params.alpha}
-      vsearch --threads ${task.cpus} --fastq_qmax ${params.maxQuality} --uchime3_denovo "${id}_centroids.fasta" --nonchimeras "${id}_zotus.fasta" --relabel Zotu 
-      vsearch --threads ${task.cpus} --fastq_qmax ${params.maxQuality} --usearch_global ${relabeled_merged} --db "${id}_zotus.fasta" --id 0.97 --otutabout zotu_table.tsv
+      vsearch \
+        --threads ${task.cpus} --fastq_qmax ${params.maxQuality} \
+        --derep_fulllength ${relabeled_merged} --sizeout \
+        --output "${id}_unique.fasta"
+      vsearch \
+        --threads ${task.cpus} --fastq_qmax ${params.maxQuality} \
+        --cluster_unoise "${id}_unique.fasta" --centroids "${id}_centroids.fasta" \
+        --minsize ${params.minAbundance} --unoise_alpha ${params.alpha}
+      vsearch \
+        --threads ${task.cpus} --fastq_qmax ${params.maxQuality} \
+        --uchime3_denovo "${id}_centroids.fasta" --nonchimeras "${id}_zotus.fasta" \
+        --relabel Zotu 
+      vsearch \
+        --threads ${task.cpus} --fastq_qmax ${params.maxQuality} \
+        --usearch_global ${relabeled_merged} --db "${id}_zotus.fasta" \
+        --id ${params.zotuIdentity} --otutabout zotu_table.tsv \
+        --userout zotu_map.tsv --userfields "query+target" \
+        --top_hits_only
     else
       >&2 echo "Merged FASTA is empty. Did your PCR primers match anything?"  
       exit 1
@@ -442,9 +457,13 @@ process dereplicate {
     # 2. run denoising & chimera removal
     # 3. generate zotu table
     if [ -s "${relabeled_merged}" ]; then
-      ${denoiser} -fastx_uniques ${relabeled_merged} -sizeout -fastaout "${id}_unique.fasta"
-      ${denoiser} -unoise3 "${id}_unique.fasta"  -zotus "${id}_zotus.fasta" -tabbedout "${id}_unique_unoise3.txt" -minsize ${params.minAbundance} -unoise_alpha ${params.alpha}
-      ${denoiser} -otutab ${relabeled_merged} -zotus ${id}_zotus.fasta -otutabout zotu_table.tsv -mapout zmap.txt
+      ${denoiser} -fastx_uniques ${relabeled_merged} \
+        -sizeout -fastaout "${id}_unique.fasta"
+      ${denoiser} -unoise3 "${id}_unique.fasta"  -zotus "${id}_zotus.fasta" \
+        -tabbedout "${id}_unique_unoise3.txt" -minsize ${params.minAbundance} \
+        -unoise_alpha ${params.alpha}
+      ${denoiser} -otutab ${relabeled_merged} -id ${params.zotuIdentity} \
+        -zotus ${id}_zotus.fasta -otutabout zotu_table.tsv -mapout zotu_map.tsv
     else
       >&2 echo "Merged FASTA is empty. Did your PCR primers match anything?"  
       exit 1