Diamond (#161)

* added diamond components

* removed quotes

Author: tiagofilipe12

changelog.md

@@ -25,6 +25,7 @@ resolution
 - Added component `unicycler`.
 - Added component `prokka`.
 - Added component `bcalm`.
+- Added component `diamond`.
 
 ### Minor/Other changes
 

docs/user/components/diamond.rst

@@ -0,0 +1,55 @@
+diamond
+=======
+
+Purpose
+-------
+
+This component performs ``blastx`` or ``blastp`` with diamond. The database
+used by diamond can be provided from the local disk or generated in the process.
+This component uses the same output type as abricate with the same blast output
+fields.
+
+.. note::
+    Software page: https://github.com/bbuchfink/diamond
+
+
+Input/Output type
+-----------------
+
+- Input type: ``Fasta``
+- Output type: None
+
+.. note::
+    The default input parameter for fasta data is ``--fasta``.
+
+Parameters
+----------
+
+- ``pathToDb``: Provide full path for the diamond database. If none is provided
+  then will try to fetch from the previous process. Default: None
+
+- ``fastaToDb``: Provide the full path for the fasta to construct a diamond
+  database. Default: None
+
+- ``blastType``: Defines the type of blast that diamond will do. Can wither be
+  blastx or blastp. Default: blastx
+
+Published results
+-----------------
+
+- ``results/annotation/diamond*``: Stores the results of the abricate screening
+  for each sample and for each specified database.
+
+Published reports
+-----------------
+
+None.
+
+Default directives
+------------------
+
+- ``diamond``:
+    - ``container``: flowcraft/diamond
+    - ``version``: 0.9.22-1
+    - ``memory``: { 4.GB * task.attempt }
+    - ``cpus``: 2

flowcraft/generator/components/annotation.py

@@ -76,7 +76,7 @@ class CardRgi(Process):
             - ``input_type``: fasta
             - ``output_type``: txt
             - ``ptype``: resistance gene detection (assembly)
-        """
+    """
 
     def __init__(self, **kwargs):
 
@@ -158,3 +158,49 @@ def __init__(self, **kwargs):
                 "version": "1.12"
             }
         }
+
+
+class Diamond(Process):
+    """diamond process for protein database queries
+
+        This process is set with:
+
+            - ``input_type``: fasta
+            - ``output_type``: None
+            - ``ptype``: post_assembly
+    """
+
+    def __init__(self, **kwargs):
+
+        super().__init__(**kwargs)
+
+        self.input_type = "fasta"
+        self.output_type = None
+
+        self.params = {
+            "pathToDb": {
+                "default": 'null',
+                "description": "Provide full path for the diamond database. "
+                               "If none is provided then will try to fetch from"
+                               " the previous process. Default: None"
+            },
+            "fastaToDb": {
+                "default": 'null',
+                "description": "Provide the full path for the fasta to "
+                               "construct a diamond database. Default: None"
+            },
+            "blastType": {
+                "default": "'blastx'",
+                "description": "Defines the type of blast that diamond will do."
+                               "Can wither be blastx or blastp. Default: blastx"
+            }
+        }
+
+        self.directives = {
+            "diamond": {
+                "container": "flowcraft/diamond",
+                "version": "0.9.22-1",
+                "memory": "{ 4.GB * task.attempt }",
+                "cpus": 2
+            }
+        }

flowcraft/generator/engine.py

@@ -65,6 +65,7 @@
         "check_coverage": readsqc.CheckCoverage,
         "chewbbaca": mlst.Chewbbaca,
         "dengue_typing": typing.DengueTyping,
+        "diamond": annotation.Diamond,
         "downsample_fastq": readsqc.DownsampleFastq,
         "fastqc": readsqc.FastQC,
         "fastqc_trimmomatic": readsqc.FastqcTrimmomatic,

flowcraft/generator/templates/diamond.nf

@@ -0,0 +1,53 @@
+// check if any of the parameters it defined before executing the process.
+if (!params.pathToDb{{ param_id }} && !params.fastaToDb{{ param_id }})
+    exit 1, "'You must specify either a pathToDb or fastaToDb parameter.'"
+// checks if both are defined and if so raises an error.
+else if (params.pathToDb{{ param_id }} && params.fastaToDb{{ param_id }})
+    exit 1, "'Both pathToDb and fastaToDb were given, choose just one.'"
+
+// list of blasts allowed for diamond
+allowedBlasts = ["blastp", "blastx"]
+// checks if blast type os defined
+if (!allowedBlasts.contains(params.blastType{{ param_id }}))
+    exit 1, "Provide a valid blast type: blastx or blastp"
+
+process diamond_{{ pid }}  {
+
+    // Send POST request to platform
+    {% include "post.txt" ignore missing %}
+
+    tag { sample_id }
+    publishDir "results/annotation/diamond_{{ pid }}/${sample_id}"
+
+    input:
+    set sample_id, file(assembly) from {{ input_channel }}
+    file pathToDb from params.pathToDb{{ param_id }} ?
+        Channel.fromPath(params.pathToDb{{ param_id }}) : Channel.value("NA")
+    file fastaToDb from params.fastaToDb{{ param_id }} ?
+        Channel.fromPath(params.fastaToDb{{ param_id }}) : Channel.value("NA")
+    val blast from params.blastType{{ param_id }}
+
+    output:
+    file "*.txt" into diamondOutputs
+    output:
+    {% with task_name="diamond"%}
+    {%- include "compiler_channels.txt" ignore missing -%}
+    {% endwith %}
+
+    script:
+    // Use database when available or otherwise use Fasta file
+    if (params.pathToDb{{ param_id }})
+        """
+        diamond ${blast} -d ${pathToDb} -q ${assembly} \
+        -o ${pathToDb}.txt -e 1E-20 -p ${task.cpus} \
+        -f 6 qseqid sseqid pident length mismatch gapopen qstart qend slen sstart send evalue bitscore
+        """
+    else if (params.fastaToDb{{ param_id }})
+        """
+        diamond makedb --in ${fastaToDb} -d ${fastaToDb}
+        diamond ${blast} -d ${fastaToDb}.dmnd -q ${assembly} \
+        -o ${fastaToDb}.txt -e 1E-20 -p ${task.cpus} \
+        -f 6 qseqid sseqid pident length mismatch gapopen qstart qend slen sstart send evalue bitscore
+        """
+
+}