Merge pull request #50 from TRON-Bioinformatics/49_add_subworkflow_sn…

…peff_custom_genome

49 add subworkflow snpeff custom genome

README.md

@@ -171,13 +171,24 @@ No additional parameter needs to be provided to use the default SARS-CoV-2 refer
 
 #### Using a custom reference genome
 
-These references can be customised to use a different SARS-CoV-2 reference or to analyse a different virus.
-Two files need to be provided:
+These references can be customised to use a different SARS-CoV-2 reference or to analyse a different virus.  
+Two files need to be provided:  
+
 - Use a custom reference genome by providing the parameter `--reference your.fasta`.
-- Gene annotation file in GFFv3 format `--gff your.gff`.
+- Gene annotation file in GFFv3 format `--gff your.gff`.  
+
+As well as an organism name for the SnpEff annotation:
+
+- Organism name to use in custom SnpEff database `--snpeff_organism`.
 
-Additionally, the FASTA needs bwa-mem2 indexes, .fai index and a .dict index.
-These indexes can be generated with the following two commands:
+**NOTE**: beware that for Nextflow to find these files the reference needs to be passed as an absolute path.
+
+When a custom reference genome is provided, the pipeline automatically generates a bwa-mem2 index
+in the case of fastq input and a custom SnpEff database for functional annotations using the steps 
+described below. The steps are listed here for documentation purposes and should __not be performed manually__.
+
+The reference genome FASTA needs to be indexed for different components of the pipeline.
+bwa-mem2 indices, .fai index and a .dict index can be generated with the following three commands:
 ```
 bwa-mem2 index reference.fasta
 samtools faidx reference.fasta
@@ -194,8 +205,6 @@ When running CoVigator you will also need to provide three parameters:
 - `--snpeff_data`: path to the SnpEff data folder
 - `--snpeff_config`: path to the SnpEff config file
 
-**NOTE**: beware that for Nextflow to find these indices the reference needs to be passed as an absolute path.
-
 **Limitations**
 
 - The SARS-CoV-2 specific annotations (ie: ConsHMM conservation and SARS-CoV-2 protein domains) will be skipped when 

bin/add_sample_to_vcf.py

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 import argparse
 import gzip
 

bin/assembly_variant_caller.py

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 import os
 from argparse import ArgumentParser
 from dataclasses import dataclass

bin/ivar2vcf.py

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 import os
 from argparse import ArgumentParser
 from typing import List

bin/phasing.py

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 import argparse
 import logging
 

bin/reference_genome.py

@@ -1,3 +1,4 @@
+#!/usr/bin/env python3
 from pysam import FastaFile
 
 

main.nf

@@ -2,7 +2,7 @@
 
 nextflow.enable.dsl = 2
 
-
+include { BWA_INDEX; SNPEFF_DATABASE } from './modules/00_prepare_annotation'
 include { READ_TRIMMING_PAIRED_END; READ_TRIMMING_SINGLE_END } from './modules/01_fastp'
 include { ALIGNMENT_PAIRED_END; ALIGNMENT_SINGLE_END } from './modules/02_bwa'
 include { BAM_PREPROCESSING; COVERAGE_ANALYSIS; PRIMER_TRIMMING_IVAR } from './modules/03_bam_preprocessing'
@@ -34,10 +34,12 @@ params.skip_normalization = false
 // references
 params.reference = false
 params.gff = false
-params.snpeff_data = false
-params.snpeff_config = false
+// These will be now generated by the subworkflow if using a custom annotation
+//params.snpeff_data = false
+//params.snpeff_config = false
 params.snpeff_organism = false
 params.primers = false
+params.different_virus = false
 
 params.output = "."
 params.min_mapping_quality = 20
@@ -88,20 +90,28 @@ if (params.reference == false) {
     skip_sarscov2_annotations = params.skip_sarscov2_annotations
 }
 else {
-    log.info "Using custom reference genome: ${params.reference}"
+    log.info "Using custom reference genome: ${params.reference}. Preparing references to be used with covigator pipeline"
     reference = params.reference    // do not put into a file as we need the indices
     gff = params.gff ? file(params.gff) : false
-    snpeff_data = params.snpeff_data
-    snpeff_config = params.snpeff_config
     snpeff_organism = params.snpeff_organism
+    snpeff_data = false
+    snpeff_config = false
     skip_sarscov2_annotations = true
 }
 
 primers = params.primers ? file(params.primers) : false
 
 skip_snpeff = false
-if (! snpeff_data || ! snpeff_config || ! snpeff_organism) {
-    log.info "Skipping SnpEff annotation as either --snpeff_data, --snpeff_config or --snpeff_organism was not provided"
+
+// This check is different for custom reference and default reference
+// For reference genome skip snpeff annotation only if the snpeff arguments are not correctly provided
+if (!params.reference && (! snpeff_data || ! snpeff_config || ! snpeff_organism)) {
+        log.info "Skipping SnpEff annotation as either --snpeff_data, --snpeff_config or --snpeff_organism was not provided"
+        skip_snpeff = true
+}
+// For a custom genome skip the snpeff annotation if params gff or snpeff organism are not provided
+else if (params.reference && (! gff || ! snpeff_organism)) {
+    log.info "Skipping SnpEff annotation as either --gff or --snpeff_organism was not provided to build database"
     skip_snpeff = true
 }
 
@@ -232,7 +242,19 @@ if (params.skip_bcftools && params.skip_gatk && params.skip_ivar && params.skip_
 
 
 workflow {
+    if (params.reference) {
+        if (! skip_snpeff && gff) {
+            SNPEFF_DATABASE(reference, gff, snpeff_organism)
+            snpeff_data = SNPEFF_DATABASE.out.snpeff_data
+            snpeff_config = SNPEFF_DATABASE.out.snpeff_config
+        }
+
+    }
     if (input_fastqs) {
+        if (params.reference) {
+            BWA_INDEX(reference)
+            reference = BWA_INDEX.out.reference
+        }
         if (library == "paired") {
             READ_TRIMMING_PAIRED_END(input_fastqs)
             ALIGNMENT_PAIRED_END(READ_TRIMMING_PAIRED_END.out[0], reference)

modules/00_prepare_annotation.nf

@@ -0,0 +1,62 @@
+params.memory = "3g"
+params.cpus = 1
+params.output = "."
+
+
+process BWA_INDEX {
+    cpus params.cpus
+    memory params.memory
+    publishDir "${params.output}", mode: "copy"
+    tag "${name}"
+
+    conda (params.enable_conda ? "bioconda::bwa-mem2=2.2.1 bioconda::samtools=1.12 bioconda::gatk4=4.2.0.0" : null)
+
+    input:
+        val(reference)
+
+    output:
+        path("reference/sequences.fa"), emit: reference
+        path("reference/sequences.fa.fai"), emit: fai
+        path("reference/sequences.dict"), emit: gatk_dict
+
+    script:
+        memory = "${params.memory}".replaceAll(" ", "").toLowerCase()
+        """
+        mkdir -p reference
+        cp ${reference} reference/sequences.fa
+        bwa-mem2 index reference/sequences.fa
+        samtools faidx reference/sequences.fa
+        gatk CreateSequenceDictionary --REFERENCE reference/sequences.fa 
+        """
+}
+
+process SNPEFF_DATABASE {
+    cpus params.cpus
+    memory params.memory
+    publishDir "${params.output}", mode: "copy"
+    tag "${name}"
+
+    conda (params.enable_conda ? "bioconda::snpeff=5.0 bioconda::samtools=1.12" : null)
+
+    input:
+        val(reference)
+        val(gff)
+        val(snpeff_organism)
+
+    output:
+        path("snpeff/snpEff.config"), emit: snpeff_config
+        path("snpeff/"), emit: snpeff_data
+
+    script:
+        memory = "${params.memory}".replaceAll(" ", "").toLowerCase()
+        """
+        mkdir -p snpeff/${snpeff_organism}
+        echo ${snpeff_organism}.genome : ${snpeff_organism} > snpeff/snpEff.config
+        cp ${reference} snpeff/${snpeff_organism}/sequences.fa
+        cp ${gff} snpeff/${snpeff_organism}/genes.gff
+        cd snpeff
+        snpEff build -gff3 -v ${snpeff_organism} -dataDir .
+        """
+
+}
+

nextflow.config

@@ -40,12 +40,12 @@ profiles {
     dag.enabled = false
   }
   test_fasta {
-    params.fasta = "$baseDir/test_data/test_data.fasta"
+    params.fasta = "$baseDir/tests/test_data/test_data.fasta"
     params.name = "test"
     params.output = "covigator_fasta_test"
   }
   test_fastq {
-    params.fastq1 = "$baseDir/test_data/test_data_1.fastq.gz"
+    params.fastq1 = "$baseDir/tests/test_data/test_data_1.fastq.gz"
     params.name = "test"
     params.output = "covigator_fastq_test"
   }