rliterman@0: // Subworkflow to run MUMmer for query/referece comparisons
rliterman@0: 
rliterman@0: // Set path variables
rliterman@0: output_directory = file(params.output_directory)
rliterman@0: mummer_directory = file(params.mummer_directory)
rliterman@0: mummer_log_directory = file(params.mummer_log_directory)
rliterman@0: snpdiffs_directory = file(params.snpdiffs_directory)
rliterman@0: log_directory = file(params.log_directory)
rliterman@0: 
rliterman@0: if(params.tmp_dir == ""){
rliterman@0:     temp_dir = ""
rliterman@0: } else{
rliterman@0:     temp_dir = file(params.temp_dir)
rliterman@0: }
rliterman@0: 
rliterman@0: ref_mode = params.ref_mode
rliterman@0: ref_id_file = file(params.ref_id_file)
rliterman@0: 
rliterman@0: // Set path to accessory scripts/files
rliterman@0: all_snpdiffs_list = file("${log_directory}/All_SNPDiffs.txt")
rliterman@0: isolate_data_file = file("${output_directory}/Isolate_Data.tsv")
rliterman@0: snpdiffs_summary_file = file("${output_directory}/Raw_MUMmer_Summary.tsv")
rliterman@0: mummerScript = file("$projectDir/bin/compileMUMmer.py")
rliterman@0: 
rliterman@0: workflow {
rliterman@0:     main:
rliterman@0:     // Align genomes
rliterman@0:     snpdiffs = alignGenomes(to_align: read_data, snpdiffs_data: snpdiffs_data)
rliterman@0:     publish:
rliterman@0:     // Publish snpdiffs
rliterman@0:     snpdiffs >> 'snpdiffs.tsv'
rliterman@0: }
rliterman@0: 
rliterman@0: workflow alignGenomes{
rliterman@0:     take:
rliterman@0:     to_align
rliterman@0:     snpdiffs_data
rliterman@0: 
rliterman@0:     emit:
rliterman@0:     return_snpdiffs
rliterman@0: 
rliterman@0:     main:
rliterman@0:     
rliterman@0:     // Align anything that needs aligning
rliterman@0:     sample_pairwise = to_align
rliterman@0:     .filter{"${it[0]}" != "${it[2]}"} // Don't map things to themselves
rliterman@0:     | runMUMmer 
rliterman@0:     | splitCsv
rliterman@0:     
rliterman@0:     log_hold = sample_pairwise
rliterman@0:     .concat(snpdiffs_data)
rliterman@0:     .unique{it -> it[2]}
rliterman@0:     .collect{it -> it[2]}
rliterman@0: 
rliterman@0:     snpdiff_files = saveMUMmerLog(log_hold)
rliterman@0:     .collect().flatten().collate(1)
rliterman@0: 
rliterman@0:     return_snpdiffs = sample_pairwise
rliterman@0:     .concat(snpdiffs_data)
rliterman@0:     .map { it -> tuple([it[0], it[1]].sort().join(',').toString(),it[0], it[1], it[2]) }
rliterman@0:     .unique{it -> it[0]}
rliterman@0:     .map{it->tuple(it[3],it[1],it[2])}
rliterman@0:     .join(snpdiff_files,by:0)
rliterman@0:     .map{it->tuple(it[1],it[2],it[0])}
rliterman@0: }
rliterman@0: 
rliterman@0: process runMUMmer{
rliterman@0: 
rliterman@15:     label 'mummerMem'
rliterman@0: 
rliterman@0:     input:
rliterman@0:     tuple val(query_name),val(query_fasta),val(ref_name),val(ref_fasta)
rliterman@0: 
rliterman@0:     output:
rliterman@0:     stdout
rliterman@0:     
rliterman@0:     script:
rliterman@0: 
rliterman@0:     report_id = "${query_name}__vs__${ref_name}"
rliterman@0:     mummer_log = file("${mummer_log_directory}/${report_id}.log")
rliterman@0: 
rliterman@0:     // Ensure MUmmer directories exist
rliterman@0:     if(!mummer_directory.isDirectory()){
rliterman@0:         error "$mummer_directory does not exist..."
rliterman@0:     } else{
rliterman@0:         """
rliterman@0:         $params.load_mummer_module
rliterman@0:         $params.load_python_module
rliterman@0:         $params.load_bedtools_module
rliterman@0:         $params.load_bbtools_module
rliterman@0: 
rliterman@0:         cd ${mummer_directory}
rliterman@0:         dnadiff -p ${report_id} ${ref_fasta} ${query_fasta}
rliterman@0:         
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.mdelta
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.mcoords
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.1delta
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.delta
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.qdiff
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.rdiff
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.unref
rliterman@0:         rm -rf ${mummer_directory}/${report_id}.unqry
rliterman@0: 
rliterman@0:         python ${mummerScript} --query "${query_name}" --query_fasta "${query_fasta}" --reference "${ref_name}" --reference_fasta "${ref_fasta}" --mummer_dir "${mummer_directory}" --snpdiffs_dir "${snpdiffs_directory}" --temp_dir "${temp_dir}" --log_file "${mummer_log}"
rliterman@0:         """
rliterman@0:     }
rliterman@0: }
rliterman@0: 
rliterman@0: process saveMUMmerLog{
rliterman@0: 
rliterman@0:     executor = 'local'
rliterman@0:     cpus = 1
rliterman@0:     maxForks = 1
rliterman@0: 
rliterman@0:     input:
rliterman@0:     val(snpdiffs_paths)
rliterman@0: 
rliterman@0:     output:
rliterman@0:     val(snpdiffs_paths)
rliterman@0: 
rliterman@0:     script:
rliterman@0:     saveSNPDiffs = file("$projectDir/bin/saveSNPDiffs.py")
rliterman@0:     all_snpdiffs_list.write(snpdiffs_paths.join('\n') + '\n')
rliterman@0:     """
rliterman@0:     $params.load_python_module
rliterman@0:     python $saveSNPDiffs --snpdiffs_file "${all_snpdiffs_list}" --summary_file "${snpdiffs_summary_file}" --isolate_file "${isolate_data_file}" --trim_name "${params.trim_name}" --ref_id_file "${ref_id_file}"
rliterman@0:     """
rliterman@0: }