rliterman@0: #! /usr/bin/env nextflow
rliterman@0: nextflow.enable.dsl=2
rliterman@0: 
rliterman@0: // CSP2 Main Script
rliterman@0: // Params are read in from command line or from nextflow.config and/or conf/profiles.config
rliterman@0: 
rliterman@0: // Check if help flag was passed
rliterman@0: help1 = "${params.help}" == "nohelp" ? "nohelp" : "help"
rliterman@0: help2 = "${params.h}" == "nohelp" ? "nohelp" : "help"
rliterman@0: 
rliterman@0: def printHelp() {
rliterman@0:     println """
rliterman@0: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
rliterman@0:     CSP2
rliterman@0: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
rliterman@0: 
rliterman@0: Global default params:
rliterman@0: 
rliterman@0:   --out            Set name for output folder/file prefixes (Default: CSP2_<timestamp>)
rliterman@0:   --outroot        Set output parent directory (Default: CWD; Useful to hardset in nextflow.config if 
rliterman@0:                    you want all output go to the same parent folder, with unique IDs set by --out)
rliterman@0:   --tmp_dir        Manually specify a TMP directory for pybedtools output
rliterman@0:   --help/--h       Display this help menu
rliterman@0: 
rliterman@0: 
rliterman@0: CSP2 can run in the following run modes:
rliterman@0: 
rliterman@0:   --runmode        Run mode for CSP2:
rliterman@0: 
rliterman@0:                    - assemble: Assemble read data (--reads/--ref_reads) into FASTA using SKESA
rliterman@0:   
rliterman@0:                    - align: Given query data (--reads/--fasta) and reference data (--ref_reads/--ref_fasta), 
rliterman@0:                             run MUMmer alignment analysis for each query/ref combination
rliterman@0:   
rliterman@0:                    - screen: Given query data (--reads/--fasta) and reference data (--ref_reads/--ref_fasta) 
rliterman@0:                              and/or MUMmer output (.snpdiffs), create a report for raw SNP 
rliterman@0:                              distances between each query and reference assembly
rliterman@0:   
rliterman@0:                    - snp: Given query data (--reads/--fasta) and reference data (--ref_reads/--ref_fasta) 
rliterman@0:                           and/or MUMmer output (.snpdiffs), generate alignments and pairwise 
rliterman@0:                           distances for all queries based on each reference dataset
rliterman@0: 
rliterman@0: Input Data:
rliterman@0: 
rliterman@0:   --fasta          Location for query isolate assembly data (.fasta/.fa/.fna). Can be a list of files, a path 
rliterman@0:                    to a signle single FASTA, or a path to a directories with assemblies. 
rliterman@0:   --ref_fasta      Location for reference isolate assembly data (.fasta/.fa/.fna). Can be a list of files, a 
rliterman@0:                    path to a signle single FASTA, or a path to a directories with assemblies. 
rliterman@0: 
rliterman@0:   --reads          Directory or list of directories containing query isolate read data
rliterman@0:   --readext        Read file extension (Default: fastq.gz)
rliterman@0:   --forward        Forward read file suffix (Default: _1.fastq.gz)
rliterman@0:   --reverse        Reverse read file suffix (Default: _2.fastq.gz)
rliterman@0:   
rliterman@0:   --ref_reads      Directory or list of directories containing reference isolate read data
rliterman@0:   --ref_readext    Reference read file extension (Default: fastq.gz)
rliterman@0:   --ref_forward    Reference forward read file suffix (Default: _1.fastq.gz)
rliterman@0:   --ref_reverse    Reference reverse read file suffix (Default: _2.fastq.gz)
rliterman@0: 
rliterman@0:   --snpdiffs       Location for pre-generated snpdiffs files (List of snpdiffs files, directory with snpdiffs)
rliterman@0: 
rliterman@0:   --ref_id         IDs to specify reference sequences (Comma-separated list; e.g., Sample_A,Sample_B,Sample_C)
rliterman@0: 
rliterman@0:   --trim_name      A common string to remove from all sample IDs (Default: ''; Useful if all assemblies end in 
rliterman@0:                    something like "_contigs_skesa.fasta")
rliterman@0: 
rliterman@0:   --n_ref          If running in --runmode snp, the number of reference genomes for CSP2 to select if none are provided (Default: 1)
rliterman@0: 
rliterman@0:   --exclude        A comma-separated list of IDs to remove prior to analysis (Useful for removing low quality 
rliterman@0:                    isolates in combination with --snpdiffs)
rliterman@0: 
rliterman@0: QC variables:
rliterman@0: 
rliterman@0:   --min_cov        Only consider queries if the reference genome is covered by at least <min_cov>% (Default: 85)
rliterman@0:   --min_len        Only consider SNPs from contig alignments longer than <min_len> bp (Default: 500)
rliterman@0:   --min_iden       Only consider SNPs from alignments with at least <min_iden> percent identity (Default: 99)
rliterman@0:   --dwin           A comma-separated set of window sizes for SNP density filters (Default: 1000,125,15; Set --dwin 0 to disable density filtering)
rliterman@0:   --wsnps          A comma-separated set of maximum SNP counts per window above (Default: 3,2,1)
rliterman@0:   --max_missing    If running in --runmode snp, mask SNPs where data is missing or purged from <max_missing>% of isolates (Default: 50)
rliterman@0: 
rliterman@0: Edge Trimming:
rliterman@0: 
rliterman@0:   --ref_edge       Don't include SNPs that fall within <ref_edge>bp of a reference contig edge (Default: 150) 
rliterman@0:   --query_edge     Don't include SNPs that fall within <query_edge>bp of a query contig edge (Default: 150)
rliterman@0:   --rescue         If flagged (Default: not flagged), sites that were filtered out due solely to query edge proximity are rescued if 
rliterman@0:                    the same reference position is covered more centrally by another query
rliterman@0: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
rliterman@0: 
rliterman@0: Example Commands: 
rliterman@0: 
rliterman@0: 1) Run CSP2 in SNP Pipeline mode using all the FASTA from /my/data/dir, and choose 3 references
rliterman@0: 
rliterman@0: nextflow run CSP2.nf --runmode snp --fasta /my/data/dir --n_ref 3
rliterman@0: 
rliterman@0: 2) Screen all the paired-end .fastq files from /my/read/dir against the reference isolate in /my/reference/isolates.txt
rliterman@0: 
rliterman@0: nextflow run CSP2.nf --runmode screen --ref_fasta /my/reference/isolates.txt --reads /my/read/dir --readext .fastq --forward _1.fastq --reverse _2.fastq
rliterman@0: 
rliterman@0: 3) Re-run the SNP pipeline using old snpdiffs files after changing the density filters and removing a bad sample
rliterman@0: 
rliterman@0: nextflow run CSP2.nf --runmode snp --snpdiffs /my/old/analysis/snpdiffs --dwin 5000,2500,1000 --wsnps 6,4,2 --ref_id Sample_A --exclude Sample_Q --out HQ_Density 
rliterman@0: 
rliterman@0: 4) Run in assembly mode and use HPC modules specified in profiles.config (NOTE: Setting the profile in nextflow uses a single hyphen (-) as compared to other arguments (--))
rliterman@0: 
rliterman@0: nextflow run CSP2.nf -profile myHPC --runmode assemble --reads /my/read/dir --out Assemblies
rliterman@0: 
rliterman@0: 5) Run in SNP pipeline mode using SLURM and use the built in conda environment (NOTE: For local jobs using conda, use -profile standard_conda)
rliterman@0: 
rliterman@0: nextflow run CSP2.nf -profile slurm_conda --runmode snp --fasta /my/data/dir --out CSP2_Conda
rliterman@0: 
rliterman@0: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
rliterman@0: """
rliterman@0:     System.exit(0)
rliterman@0: }
rliterman@0: 
rliterman@0: if (help1 == "help") {
rliterman@0:     printHelp()
rliterman@0: } else if(help2 =="help"){
rliterman@0:     printHelp()
rliterman@0: }
rliterman@0: 
rliterman@0: // Assess run mode
rliterman@0: if (params.runmode == "") {
rliterman@0:     error "--runmode must be specified..."
rliterman@0: } else if (!['align','assemble', 'screen', 'snp','conda_init'].contains(params.runmode)){
rliterman@0:     error "--runmode must be 'align','assemble', 'screen', or 'snp', not ${params.runmode}..."
rliterman@0: }
rliterman@0: 
rliterman@0: // If runmode is conda_init, launch a local process to spurn the generation of the conda environment and exit
rliterman@0: if (params.runmode != "conda_init") {
rliterman@0: 
rliterman@0:     // Ensure necessary data is provided given the run mode
rliterman@0:     // Runmode 'assemble'
rliterman@0:     //  - Requires: --reads/--ref_reads
rliterman@0:     //  - Runs SKESA and summarzies output FASTA 
rliterman@0:     if (params.runmode == "assemble"){
rliterman@0:         if((params.reads == "") && (params.ref_reads == "")){
rliterman@0:             error "Runmode is --assemble but no read data provided via --reads/--ref_reads"
rliterman@0:         } 
rliterman@0:     }
rliterman@0: 
rliterman@0:     // Runmode 'align'
rliterman@0:     //  - Requires: --reads/--fasta/--snpdiffs
rliterman@0:     //  - Optional: --ref_reads/--ref_fasta/--ref_id
rliterman@0:     //  - Runs MUMmer, generates .snpdiffs, and alignment summary.
rliterman@0:     //      - If references are provided via --ref_reads/--ref_fasta/--ref_id, non-reference samples are aligned to each reference
rliterman@0:     //      - If no references are provided, alignments are all-vs-all
rliterman@0:     //      - If --snpdiffs are provided, their FASTAs will be autodetected and, if present, used as queries or references as specified by --ref_reads/--ref_fasta/--ref_id
rliterman@0:     //      - Does NOT perform QC filtering
rliterman@0: 
rliterman@0:     else if (params.runmode == "align"){
rliterman@0:         if((params.fasta == "") && (params.reads == "") && (params.snpdiffs == "")){
rliterman@0:             error "Runmode is --align but no query data provided via --fasta/--reads/--snpdiffs"
rliterman@0:         } 
rliterman@0:     }
rliterman@0: 
rliterman@0:     // Runmode 'screen'
rliterman@0:     //  - Requires: --reads/--fasta/--snpdiffs
rliterman@0:     //  - Optional: --ref_reads/--ref_fasta/--ref_id
rliterman@0:     //  - Generates .snpdiffs files (if needed), applies QC, and generates alignment summaries and SNP distance estimates
rliterman@0:     //      - If references are provided via --ref_reads/--ref_fasta/--ref_id, non-reference samples are aligned to each reference
rliterman@0:     //      - If no references are provided, alignments are all-vs-all
rliterman@0:     //      - If --snpdiffs are provided, (1) they will be QC filtered and included in the output report and (2) their FASTAs will be autodetected and, if present, used as queries or references as specified by --ref_reads/--ref_fasta/--ref_id
rliterman@0: 
rliterman@0:     else if (params.runmode == "screen"){
rliterman@0:         if((params.fasta == "") && (params.reads == "") && (params.snpdiffs == "")){
rliterman@0:             error "Runmode is --screen but no query data provided via --snpdiffs/--reads/--fasta"
rliterman@0:         }
rliterman@0:     }
rliterman@0: 
rliterman@0:     // Runmode 'snp'
rliterman@0:     //  - Requires: --reads/--fasta/--snpdiffs
rliterman@0:     //  - Optional: --ref_reads/--ref_fasta/--ref_id
rliterman@0:     //  - If references are not provided, runs RefChooser using all FASTAs to choose references (--n_ref sets how many references to choose)
rliterman@0:     //  - Each query is aligned to each reference, and pairwise SNP distances for all queries are generated based on that reference
rliterman@0:     //  - Generates .snpdiffs files (if needed), applies QC, and generates SNP distance data between all queries based on their alignment to each reference
rliterman@0:     else if (params.runmode == "snp"){
rliterman@0:         if((params.snpdiffs == "") && (params.fasta == "") && (params.reads == "")) {
rliterman@0:             error "Runmode is --snp but no query data provided via --snpdiffs/--reads/--fasta"
rliterman@0:         }
rliterman@0:     } 
rliterman@0: 
rliterman@0:     // Set directory structure
rliterman@0:     if (params.outroot == "") {
rliterman@0:         output_directory = file(params.out)
rliterman@0:     } else {
rliterman@0:         out_root = file(params.outroot)
rliterman@0:         output_directory = file("${out_root}/${params.out}")
rliterman@0:     }
rliterman@0: 
rliterman@0:     // If the output directory exists, create a new subdirectory with the default output name ("CSP2_<TIME>")
rliterman@0:     if(!output_directory.getParent().isDirectory()){
rliterman@0:         error "Parent directory for output (--outroot) is not a valid directory [${output_directory.getParent()}]..."
rliterman@0:     } else if(output_directory.isDirectory()){
rliterman@0:         output_directory = file("${output_directory}/CSP2_${new java.util.Date().getTime()}")
rliterman@0:         output_directory.mkdirs()
rliterman@0:     } else{
rliterman@0:         output_directory.mkdirs()
rliterman@0:     }
rliterman@0: 
rliterman@0:     if(params.tmp_dir != ""){
rliterman@0:         tempdir = file(params.tmp_dir)
rliterman@0: 
rliterman@0:         if(tempdir.isDirectory()){
rliterman@0:             temp_dir = file("${tempdir}/CSP2_${new java.util.Date().getTime()}_tmp")
rliterman@0:             temp_dir.mkdirs()
rliterman@0:             params.temp_dir = file(temp_dir)
rliterman@0: 
rliterman@0:         } else if(tempdir.getParent().isDirectory()){
rliterman@0:             tempdir.mkdirs()
rliterman@0:             params.temp_dir = tempdir
rliterman@0:         } else{
rliterman@0:             error "Parent directory for temp directory --tmp_dir (${params.tmp_dir}) is not a valid directory..."
rliterman@0:         }
rliterman@0:     } else{
rliterman@0:         params.temp_dir = ""
rliterman@0:     }
rliterman@0: 
rliterman@0:     // Set MUMmer and SNP directories
rliterman@0:     mummer_directory = file("${output_directory}/MUMmer_Output")
rliterman@0:     snpdiffs_directory = file("${output_directory}/snpdiffs")
rliterman@0:     snp_directory = file("${output_directory}/SNP_Analysis")
rliterman@0: 
rliterman@0:     // Set paths for output files 
rliterman@0:     isolate_data_file = file("${output_directory}/Isolate_Data.tsv")
rliterman@0:     screening_results_file = file("${output_directory}/Screening_Results.tsv")
rliterman@0: 
rliterman@0:     // In --runmode assembly, results save to output_directory
rliterman@0:     if(params.runmode == "assemble"){
rliterman@0:         ref_mode = false
rliterman@0: 
rliterman@0:         log_directory = file("${output_directory}")
rliterman@0:         assembly_directory = file("${output_directory}")
rliterman@0: 
rliterman@0:         // Set dummy paths for log files
rliterman@0:         screen_log_dir = file("${log_directory}/Screening_Logs")
rliterman@0:         snp_log_dir = file("${log_directory}/SNP_Logs")
rliterman@0:         ref_id_file = file("${log_directory}/Reference_IDs.txt")
rliterman@0:         mummer_log_directory = file("${log_directory}/MUMmer_Logs")
rliterman@0:         mash_dir = file("${log_directory}/sketch_dir")
rliterman@0: 
rliterman@0:     } else{
rliterman@0: 
rliterman@0:         // Get reference mode
rliterman@0:         if(params.ref_reads == "" && params.ref_fasta == "" && params.ref_id == ""){
rliterman@0:             ref_mode = false
rliterman@0:         } else{
rliterman@0:             ref_mode = true
rliterman@0:         }
rliterman@0:         
rliterman@0:         // Set directories
rliterman@0:         log_directory = file("${output_directory}/logs")
rliterman@0:         assembly_directory = file("${output_directory}/Assemblies")
rliterman@0:         mummer_log_directory = file("${log_directory}/MUMmer_Logs")
rliterman@0:         ref_id_file = file("${log_directory}/Reference_IDs.txt")
rliterman@0: 
rliterman@0:         // Create directories
rliterman@0:         log_directory.mkdirs()
rliterman@0:         mummer_directory.mkdirs()
rliterman@0:         mummer_log_directory.mkdirs()
rliterman@0:         snpdiffs_directory.mkdirs()
rliterman@0: 
rliterman@0:         // Touch Reference_IDs.txt to establish it
rliterman@0:         file(ref_id_file).text = ''
rliterman@0: 
rliterman@0:         // Set paths for log subdirectories
rliterman@0:         screen_log_dir = file("${log_directory}/Screening_Logs")
rliterman@0:         snp_log_dir = file("${log_directory}/SNP_Logs")
rliterman@0:         mash_dir = file("${log_directory}/sketch_dir")
rliterman@0: 
rliterman@0:         // If --reads/--ref_reads are provided, prepare a directory for assemblies
rliterman@0:         if((params.reads != "") || (params.ref_reads != "")){
rliterman@0:             assembly_directory.mkdirs()
rliterman@0:         }
rliterman@0: 
rliterman@0:         // If runmode is snp, prepare a directory for SNP analysis + logs
rliterman@0:         if(params.runmode == "snp"){
rliterman@0:             snp_directory.mkdirs()
rliterman@0:             snp_log_dir.mkdirs()
rliterman@0:             if(!ref_mode){
rliterman@0:                 mash_dir.mkdirs()
rliterman@0:             }        
rliterman@0:         }
rliterman@0: 
rliterman@0:         // If runmode is screen, prepare a directory for screening logs
rliterman@0:         if(params.runmode == "screen"){
rliterman@0:             screen_log_dir.mkdirs()
rliterman@0:         }
rliterman@0:     }
rliterman@0: 
rliterman@0:     // Parameterize variables to pass between scripts
rliterman@0:     params.output_directory = file(output_directory)
rliterman@0:     params.log_directory = file(log_directory)
rliterman@0:     params.screen_log_dir = file(screen_log_dir)
rliterman@0:     params.snp_log_dir = file(snp_log_dir)
rliterman@0:     params.assembly_directory = file(assembly_directory)
rliterman@0:     params.mummer_directory = file(mummer_directory)
rliterman@0:     params.mummer_log_directory = file(mummer_log_directory)
rliterman@0:     params.snpdiffs_directory = file(snpdiffs_directory)
rliterman@0:     params.snp_directory = file(snp_directory)
rliterman@0:     params.ref_id_file = file(ref_id_file)
rliterman@0:     params.mash_directory = file(mash_dir)
rliterman@0: 
rliterman@0:     params.ref_mode = ref_mode
rliterman@0: 
rliterman@0:     // Set up modules if needed
rliterman@0:     params.load_python_module = params.python_module == "" ? "" : "module load -s ${params.python_module}"
rliterman@0:     params.load_skesa_module = params.skesa_module == "" ? "" : "module load -s ${params.skesa_module}"
rliterman@0:     params.load_bedtools_module = params.bedtools_module == "" ? "" : "module load -s ${params.bedtools_module}"
rliterman@0:     params.load_bbtools_module = params.bbtools_module == "" ? "" : "module load -s ${params.bbtools_module}"
rliterman@0:     params.load_mummer_module = params.mummer_module == "" ? "" : "module load -s ${params.mummer_module}"
rliterman@0:     params.load_mash_module = params.mash_module == "" ? "" : "module load -s ${params.mash_module}"
rliterman@0: 
rliterman@0:     // Save params to log file
rliterman@0:     params.each { key, value ->
rliterman@0:         file("${log_directory}/CSP2_Params.txt") << "$key = $value\n"
rliterman@0:     }
rliterman@0: } else{
rliterman@0:     params.output_directory = "./"
rliterman@0:     params.log_directory = "./"
rliterman@0:     params.screen_log_dir = "./"
rliterman@0:     params.snp_log_dir = "./"
rliterman@0:     params.assembly_directory = "./"
rliterman@0:     params.mummer_directory = "./"
rliterman@0:     params.mummer_log_directory = "./"
rliterman@0:     params.snpdiffs_directory = "./"
rliterman@0:     params.snp_directory = "./"
rliterman@0:     params.ref_id_file = "./"
rliterman@0:     params.mash_directory = "./"
rliterman@0:     params.ref_mode = false
rliterman@0:     params.load_python_module = ""
rliterman@0:     params.load_skesa_module = ""
rliterman@0:     params.load_bedtools_module = ""
rliterman@0:     params.load_bbtools_module = ""
rliterman@0:     params.load_mummer_module = ""
rliterman@0:     params.load_mash_module = ""
rliterman@0: }
rliterman@0: 
rliterman@0: //////////////////////////////////////////////////////////////////////////////////////////
rliterman@0: 
rliterman@0: // Import modules
rliterman@0: include {fetchData} from "./subworkflows/fetchData/main.nf"
rliterman@0: include {alignGenomes} from "./subworkflows/alignData/main.nf"
rliterman@0: include {runScreen;runSNPPipeline} from "./subworkflows/snpdiffs/main.nf"
rliterman@0: include {runRefChooser} from "./subworkflows/refchooser/main.nf"
rliterman@0: 
rliterman@0: workflow{
rliterman@0:     
rliterman@0:     if(params.runmode == "conda_init"){
rliterman@0:         conda_init()
rliterman@0:     } else{
rliterman@0:         // Read in data
rliterman@0:         input_data = fetchData()
rliterman@0: 
rliterman@0:         query_data = input_data.query_data
rliterman@0:         reference_data = input_data.reference_data
rliterman@0:         snpdiffs_data = input_data.snpdiff_data
rliterman@0: 
rliterman@0:         // Create channel for pre-aligned data [(Query_1,Query_2),SNPDiffs_File]
rliterman@0:         prealigned = snpdiffs_data
rliterman@0:         .map { it -> tuple([it[0], it[1]].sort().join(',').toString(), it[2]) }
rliterman@0:         .unique{it -> it[0]}
rliterman@0:         
rliterman@0:         // If run mode is 'assemble', tasks are complete
rliterman@0:         if((params.runmode == "align") || (params.runmode == "screen") || (params.runmode == "snp")){
rliterman@0: 
rliterman@0:             // If there is no reference data, align all query_data against each other
rliterman@0:             if(!ref_mode){
rliterman@0: 
rliterman@0:                 if((params.runmode == "align") || (params.runmode == "screen")){
rliterman@0:                     seen_combinations = []
rliterman@0:                     
rliterman@0:                     to_align = query_data.combine(query_data) // Self-combine query data
rliterman@0:                     .collect().flatten().collate(4)
rliterman@0:                     .filter{it -> (it[1].toString() != "null") && (it[3].toString() != "null")} // Can't align without FASTA
rliterman@0:                     .filter{ it -> // Get unique combinations
rliterman@0:             
rliterman@0:                     combination = ["${it[0]}", "${it[2]}"].sort()
rliterman@0:                     
rliterman@0:                     if(combination in seen_combinations) {
rliterman@0:                         return false
rliterman@0:                     } else {
rliterman@0:                         seen_combinations << combination
rliterman@0:                         return true
rliterman@0:                     }}
rliterman@0:                 } 
rliterman@0:                 
rliterman@0:                 // If running SNP pipeline without references, run RefChooser to choose references
rliterman@0:                 else if(params.runmode == "snp"){
rliterman@0:                     reference_data = runRefChooser(query_data)
rliterman@0:                     
rliterman@0:                     to_align = query_data
rliterman@0:                     .combine(reference_data)
rliterman@0:                     .filter{it -> (it[1].toString() != "null") && (it[3].toString() != "null")} // Can't align without FASTA
rliterman@0:                 }
rliterman@0:             }
rliterman@0: 
rliterman@0:             // If references are provided, align all queries against all references
rliterman@0:             else{
rliterman@0:                 to_align = query_data
rliterman@0:                 .combine(reference_data)
rliterman@0:                 .filter{it -> (it[1].toString() != "null") && (it[3].toString() != "null")} // Can't align without FASTA
rliterman@0:             }
rliterman@0: 
rliterman@0:             // Don't align things that are already aligned via --snpdiffs
rliterman@0:             unaligned = to_align
rliterman@0:             .map { it -> tuple([it[0], it[2]].sort().join(',').toString(),it[0], it[1], it[2], it[3]) }
rliterman@0:             .unique{it -> it[0]}
rliterman@0:             .join(prealigned, by:0, remainder:true)
rliterman@0:             .filter{it -> it[5].toString() == "null"}
rliterman@0:             .map{it -> [it[1], it[2], it[3], it[4]]}
rliterman@0:             | collect | flatten | collate(4)
rliterman@0: 
rliterman@0:             all_snpdiffs = alignGenomes(unaligned,snpdiffs_data)
rliterman@0:             .ifEmpty { error "No .snpdiffs to process..." }
rliterman@0:             .collect().flatten().collate(3)
rliterman@0:             
rliterman@0:             if(params.runmode == "screen"){
rliterman@0:                 runScreen(all_snpdiffs)
rliterman@0:             } else if(params.runmode == "snp"){
rliterman@0:                 runSNPPipeline(all_snpdiffs,reference_data)
rliterman@0:             }
rliterman@0:         }
rliterman@0:     }
rliterman@0: }
rliterman@0: 
rliterman@0: // Dummy process to stimulate conda env generation
rliterman@0: process conda_init {
rliterman@0:     executor = 'local'
rliterman@0:     cpus = 1
rliterman@0:     maxForks = 1
rliterman@0:     
rliterman@0:     script:
rliterman@0:     """
rliterman@0:     """
rliterman@0: }