Prokka (#159)

* added experimental implementation for prokka

* added documentation to the additions to prokka

* removed duplicate in changelog

Author: tiagofilipe12

changelog.md

@@ -22,6 +22,7 @@ resolution
 - Added component `abyss`.
 - Added component `bandage`.
 - Added component `unicycler`.
+- Added component `prokka`.
 
 ### Minor/Other changes
 

docs/user/components/mash_screen.rst

@@ -1,11 +1,11 @@
 mash_screen
-==============
+===========
 
 Purpose
 -------
 
 This component performes mash screen to find plasmids
-contained in high throughoput sequencing data, using as inputs read files
+contained in high throughout sequencing data, using as inputs read files
 (FastQ files). Then, the resulting file can
 be imported into `pATLAS <http://www.patlas.site/>`_.
 This component searches for containment of a given sequence in read sequencing

docs/user/components/prokka.rst

@@ -1,4 +1,54 @@
 prokka
 ======
 
-**Building...**
+
+Purpose
+-------
+
+This component performs annotations using the annotations available in
+`prokka <https://github.com/tseemann/prokka>`_.
+
+
+Input/Output type
+-----------------
+
+- Input type: ``fasta``
+- Output type: ``None``
+
+.. note::
+    - Although the component doesn't have an output channel it writes the results into the ``publishDir``.
+
+
+Parameters
+----------
+
+- ``centre``: sets the center to which the sequencing center id.
+  Default: 'UMMI'.
+
+- ``kingdom``: Selects the annotation mode between Archaea, Bacteria,
+  Mitochondria, Viruses. Default: Bacteria).
+
+- ``genus``: Allows user to select a genus name. Default: 'Genus' (same
+  as prokka). This also adds the use of the --usegenus flag to prokka.
+
+
+Published results
+-----------------
+
+- ``results/annotation/prokka_<pid>/<sample_id>``: All the outputs from prokka
+  will be available in these directories.
+
+
+Published reports
+-----------------
+
+None.
+
+
+Default directives
+------------------
+
+- ``prokka``:
+    - ``cpus``: 2
+    - ``container``: ummidock/prokka
+    - ``version``: 1.12

flowcraft/generator/components/annotation.py

@@ -134,10 +134,27 @@ def __init__(self, **kwargs):
         self.link_end.append({"link": "MAIN_assembly",
                               "alias": "MAIN_assembly"})
 
+        self.params = {
+            "centre": {
+                "default": "'UMMI'",
+                "description": "sequencing centre ID"
+            },
+            "kingdom": {
+                "default": "'Bacteria'",
+                "description": "Annotation mode: Archaea|Bacteria|Mitochondria"
+                               "|Viruses (default 'Bacteria')"
+            },
+            "genus": {
+                "default": "false",
+                "description": "Genus name (default 'Genus'). This also adds"
+                               "the --usegenus flag to prokka"
+            },
+        }
+
         self.directives = {
             "prokka": {
                 "cpus": 2,
-                "container": "ummidock/prokka-nf",
-                "version": "1.12.0-2"
+                "container": "ummidock/prokka",
+                "version": "1.12"
             }
         }

flowcraft/generator/engine.py

@@ -90,8 +90,8 @@
         "pilon": ap.Pilon,
         "process_skesa": ap.ProcessSkesa,
         "process_spades": ap.ProcessSpades,
-        "progressive_mauve":alignment.ProgressiveMauve,
-        #"prokka": annotation.Prokka,
+        "progressive_mauve": alignment.ProgressiveMauve,
+        "prokka": annotation.Prokka,
         "quast": ap.Quast,
         "raxml": phylogeny.Raxml,
         "reads_download": downloads.DownloadReads,

flowcraft/generator/templates/prokka.nf

@@ -1,4 +1,11 @@
 
+IN_centre_{{ pid }} = Channel.value(params.centre{{ param_id }})
+
+IN_kingdom_{{ pid }} = Channel.value(params.kingdom{{ param_id }})
+
+// check if genus is provided or not
+genusVar = (params.genus{{ param_id }} == false) ? "" : "--usegenus --genus ${params.genus{{param_id}}} "
+
 process prokka_{{ pid }} {
 
     // Send POST request to platform
@@ -9,6 +16,8 @@ process prokka_{{ pid }} {
 
     input:
     set sample_id, file(assembly) from {{ input_channel }}
+    val centre from IN_centre_{{ pid }}
+    val kingdom from IN_kingdom_{{ pid }}
 
     output:
     file "${sample_id}/*"
@@ -19,8 +28,8 @@ process prokka_{{ pid }} {
     script:
     """
     {
-        prokka --outdir $sample_id --cpus $task.cpus --centre UMMI --compliant \
-               --increment 10 $assembly >> .command.log 2>&1
+        prokka --outdir $sample_id --cpus $task.cpus --centre ${centre} \
+        --compliant --kingdom ${kingdom} ${genusVar} --increment 10 $assembly
         echo pass > .status
     } || {
         echo fail > .status