updating docker containers and setting templates

modules/local/CNAqc/main.nf

@@ -1,7 +1,9 @@
 process CNAQC {
     tag "$meta.id"
     label "process_low"
-    container = 'docker://lvaleriani/cnaqc:version1.0'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://lvaleriani/cnaqc:version1.0' :
+        'docker.io/lvaleriani/cnaqc:version1.0' }"
 
     input:
     tuple val(meta), path(cna_rds), path(snv_rds)
@@ -13,7 +15,6 @@ process CNAQC {
     tuple val(meta), path("*_qc.pdf"),                                  emit: plot_pdf_qc
     path "versions.yml",                                                emit: versions
 
-
     script:
     def args                                = task.ext.args
     def prefix                              = task.ext.prefix                                       ?: "${meta.id}"

modules/local/CNAqc2tsv/main.nf

@@ -1,11 +1,9 @@
-//
-// Mutations extraction from mCNAqc
-//
-
 process RDS_PROCESSING {
     tag "$meta.id"
     label "process_single"
-    container = 'docker://lvaleriani/cnaqc:version1.0'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://lvaleriani/cnaqc:version1.0' :
+        'docker.io/lvaleriani/cnaqc:version1.0' }"
 
     input:
     tuple val(meta), path(rds_join), val(tumour_samples)
@@ -14,41 +12,6 @@ process RDS_PROCESSING {
     tuple val(meta), path("*_joint_table.tsv"), val(tumour_samples),    emit: tsv
     path "versions.yml",                                                emit: versions
 
-
     script:
-    def args    = task.ext.args     ?:  ''
-    def prefix  = task.ext.prefix   ?:  "${meta.id}"
-
-    """
-    #!/usr/bin/env Rscript
-    library(dplyr)
-    library(CNAqc)
-
-    source("$moduleDir/utils.R")
-
-    multi_cnaqc = readRDS(file = "$rds_join")
-    mutations_multisample <- get_sample(m_cnaqc_obj = multi_cnaqc,sample = get_sample_name(multi_cnaqc),
-                                    which_obj = "original")
-    multisample_jointTable <- list()
-
-    for (s in get_sample_name(multi_cnaqc)){
-        purity <- mutations_multisample[[s]][["purity"]]
-        multisample_jointTable[[s]] <- mutations_multisample[[s]][["mutations"]] %>%
-            dplyr::mutate(purity=purity)
-    }
-
-    joint_table <- bind_rows(multisample_jointTable)
-
-    write.table(joint_table, file = paste0("$prefix","_joint_table.tsv"), append = F, quote = F, sep = "\t", row.names = FALSE)
-
-        # version export
-    f <- file("versions.yml","w")
-    dplyr_version <- sessionInfo()\$otherPkgs\$dplyr\$Version
-    cnaqc_version <- sessionInfo()\$otherPkgs\$CNAqc\$Version
-    writeLines(paste0('"', "$task.process", '"', ":"), f)
-    writeLines(paste("    dplyr:", dplyr_version), f)
-    writeLines(paste("    CNAqc:", cnaqc_version), f)
-    close(f)
-
-    """
+    template "main_script.R"
 }

modules/local/SigProfiler/SigProfiler/main.nf

@@ -1,7 +1,9 @@
 process SIGPROFILER {
     tag "$meta.id"
     label "process_high"
-    container = 'docker://katiad/sigprofiler:version1.0'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://katiad/sigprofiler:version1.0' :
+        'docker.io/katiad/sigprofiler:version1.0' }"
 
     input:
         tuple val(meta), path(tsv_list, stageAs: '*.tsv')
@@ -61,7 +63,6 @@ process SIGPROFILER {
     from SigProfilerExtractor import sigpro as sig
     from SigProfilerMatrixGenerator.scripts import SigProfilerMatrixGeneratorFunc as matGen
 
-
     if __name__ == '__main__':
 
         dataset_id = "$meta.dataset"
@@ -72,9 +73,7 @@ process SIGPROFILER {
 
         output_path = os.path.join("output", "SBS", f"{dataset_id}.SBS96.all")
 
-
         # input data preprocessing
-
         def process_tsv_join(tsv_list):
             patients_tsv = tsv_list.split()
             # Read each file into a pandas DataFrame and ensure all columns are of type 'string'

modules/local/SigProfiler/download/main.nf

@@ -1,6 +1,8 @@
 process DOWNLOAD_GENOME_SIGPROFILER {
     label "process_single"
-    container = 'docker://katiad/sigprofiler:version1.0'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://katiad/sigprofiler:version1.0' :
+        'docker.io/katiad/sigprofiler:version1.0' }"
 
     input:
         val(reference_genome) // reference_genome : genome -> for example: GRCh37
@@ -9,7 +11,6 @@ process DOWNLOAD_GENOME_SIGPROFILER {
         path("*"), emit: genome_sigprofiler
         path "versions.yml", emit: versions
 
-
     script:
     """
     SigProfilerMatrixGenerator install $reference_genome -v .

modules/local/SparseSignatures/main.nf

@@ -1,10 +1,11 @@
 process SPARSE_SIGNATURES {
     tag "$meta.id"
     label "process_low_long"
-    container = 'docker://lvaleriani/sparsesignature:version1.0'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://lvaleriani/sparsesignature:version1.0' :
+        'docker.io/lvaleriani/sparsesignature:version1.0' }"
 
     input:
-
         tuple val(meta), path(tsv_join,  stageAs: '*.tsv')
 
     output:
@@ -16,186 +17,5 @@ process SPARSE_SIGNATURES {
         path "versions.yml", emit: versions
 
     script:
-    def args                         = task.ext.args
-    def prefix                       = task.ext.prefix                                    ?: "${meta.id}"
-    def K                            = args!='' && args.K                                 ? "$args.K" : ""
-    def background_signature         = args!='' && args.background_signature              ? "$args.background_signature" : ""
-    def beta                         = args!='' && args.beta                              ? "$args.beta" : ""
-    def normalize_counts             = args!='' && args.normalize_counts                  ? "$args.normalize_counts" : ""
-    def nmf_runs                     = args!='' && args.nmf_runs                          ? "$args.nmf_runs" : ""
-    def iterations                   = args!='' && args.iterations                        ? "$args.iterations" : ""
-    def max_iterations_lasso         = args!='' && args.max_iterations_lasso              ? "$args.max_iterations_lasso" : ""
-    def num_processes                = args!='' && args.num_processes                     ? "$args.num_processes" : ""
-    def cross_validation_entries     = args!='' && args.cross_validation_entries          ? "$args.cross_validation_entries" : ""
-    def cross_validation_repetitions = args!='' && args.cross_validation_repetitions      ? "$args.cross_validation_repetitions" : ""
-    def cross_validation_iterations  = args!='' && args.cross_validation_iterations       ? "$args.cross_validation_iterations" : ""
-    def lambda_values_alpha          = args!='' && args.lambda_values_alpha               ? "$args.lambda_values_alpha" : ""
-    def lambda_values_beta           = args!='' && args.lambda_values_beta                ? "$args.lambda_values_beta" : ""
-    def lambda_rate_alpha            = args!='' && args.lambda_rate_alpha                 ? "$args.lambda_rate_alpha" : ""
-    def verbose                      = args!='' && args.verbose                           ? "$args.verbose" : ""
-    def seed                         = args!='' && args.seed                              ? "$args.seed" : ""
-
-    """
-    #!/usr/bin/env Rscript
-
-    library(SparseSignatures)
-    library(ggplot2)
-    library(stringr)
-    library(patchwork)
-    library(dplyr)
-
-    source("$moduleDir/getters.R")
-
-    n_procs = parse(text="$num_processes")
-        if (n_procs == "all"){
-            n_procs = as.double("Inf")
-        } else {
-            n_procs = eval(n_procs)
-    }
-
-
-    patients_tsv = strsplit("$tsv_join", " ")[[1]]
-    tables = lapply(patients_tsv, FUN = function(p_table){
-        read.delim(p_table, sep = "\\t", header=T) %>%
-            mutate(across(everything(), as.character))
-    }
-    )
-    multisample_table = dplyr::bind_rows(tables)
-
-
-    #Extract input data information
-    input_data <- multisample_table[,c("Indiv","chr","from","to","ref","alt")]
-    input_data <- setNames(input_data, c("sample","chrom","start","end","ref","alt"))
-    input_data[["end"]] <- input_data[["start"]]
-    input_data <- input_data %>% mutate(start = as.integer(start), end = as.integer(end))
-
-    #Generate the patient vs mutation count matrix from mutation data
-    #Load reference human-genome specification.
-    #The user must select, among the available choices, the reference genome consistent with the mutation dataset.
-
-    load_genome <- function(genome, input_data) {
-        if (genome == "GRCh37") {
-            library(BSgenome.Hsapiens.1000genomes.hs37d5)
-            bsg <- BSgenome.Hsapiens.1000genomes.hs37d5::hs37d5
-            input_data[["chrom"]] <- substring(input_data[["chrom"]], 4, 5)
-
-        } else if (genome == "GRCh38") {
-            library(BSgenome.Hsapiens.UCSC.hg38)
-            bsg <- BSgenome.Hsapiens.UCSC.hg38
-
-            # Leave 'chrom' unchanged for GRCh38
-        }
-        return(list(bsg = bsg, input_data = input_data))
-    }
-
-    data_list = load_genome("$params.genome", input_data)
-    bsg <- data_list[["bsg"]]
-    input_data <- data_list[["input_data"]]
-
-
-    mut_counts = SparseSignatures::import.trinucleotides.counts(data=input_data, reference=bsg)
-
-    #Load a reference SBS5 background signature from COSMIC
-    data(background)
-
-    #Estimate the initial values of beta
-    starting_betas = SparseSignatures::startingBetaEstimation(x = mut_counts,
-                                                            K = eval(parse(text="$K")),
-                                                            background_signature  = background)
-
-
-    #Find the optimal number of signatures and sparsity level: rely on cross-validation
-    # higher number of CV repetitions corresponds to more accurate parameter estimates
-
-    cv_out = SparseSignatures::nmfLassoCV(
-        x = mut_counts,
-        K = eval(parse(text="$K")),
-        starting_beta = starting_betas,
-        background_signature = background,
-        normalize_counts = as.logical("$normalize_counts"),
-        nmf_runs = as.integer("$nmf_runs"),
-        lambda_values_alpha = eval(parse(text="$lambda_values_alpha")),
-        lambda_values_beta = eval(parse(text="$lambda_values_beta")),
-        cross_validation_entries = as.numeric("$cross_validation_entries"),
-        cross_validation_iterations = as.integer("$cross_validation_iterations"),
-        cross_validation_repetitions = as.integer("$cross_validation_repetitions"),
-        iterations = as.integer("$iterations"),
-        max_iterations_lasso = as.integer("$max_iterations_lasso"),
-        num_processes = n_procs,
-        verbose = as.logical("$verbose"),
-        seed = as.integer("$seed")
-    )
-
-
-    #Analyze the mean squared error results averaging over cross-validation repetitions
-    cv_mses <- cv_out[["grid_search_mse"]][1,,]
-    cv_means_mse <- matrix(sapply(cv_mses, FUN = mean),
-                        nrow = dim(cv_mses)[1]
-    )
-
-    dimnames(cv_means_mse) <- dimnames(cv_mses)
-
-    #Find the combination of parameters that yields the lowest MSE
-    min_ii <- which(cv_means_mse == min(cv_means_mse, na.rm = TRUE), arr.ind = TRUE)
-    min_Lambda_beta <- rownames(cv_means_mse)[min_ii[1]]
-    min_Lambda_beta <- as.numeric(gsub("_Lambda_Beta", "", min_Lambda_beta))
-    min_K <- colnames(cv_means_mse)[min_ii[2]]
-    min_K <- as.numeric(gsub("_Signatures", "", min_K))
-    best_params_config <- data.frame(min_K, min_Lambda_beta)
-
-    saveRDS(object = cv_means_mse, file = paste0("$prefix", "_cv_means_mse.rds"))
-    saveRDS(object = best_params_config, file = paste0("$prefix", "_best_params_config.rds"))
-
-    #Discovering the signatures within the dataset: NMF Lasso
-    #Compute the signatures for the best configuration.
-
-    nmf_Lasso_out = SparseSignatures::nmfLasso(
-        x = mut_counts,
-        K = min_K,
-        beta = eval(parse(text="$beta")),
-        background_signature = background,
-        normalize_counts = as.logical("$normalize_counts"),
-        lambda_rate_alpha = eval(parse(text="$lambda_rate_alpha")),
-        lambda_rate_beta = min_Lambda_beta,
-        iterations = as.integer("$iterations"),
-        max_iterations_lasso = as.integer("$max_iterations_lasso"),
-        verbose = as.logical("$verbose")
-    )
-
-    saveRDS(object = nmf_Lasso_out, file =  paste0("$prefix", "_nmf_Lasso_out.rds"))
-
-    #Signature visualization
-    signatures = nmf_Lasso_out\$beta
-    plot_signatures <- SparseSignatures::signatures.plot(beta=signatures, xlabels=FALSE)
-
-    plot_exposure = nmf_Lasso_out\$alpha %>%
-        as.data.frame() %>%
-        tibble::rownames_to_column(var="PatientID") %>%
-        tidyr::pivot_longer(cols=!"PatientID", names_to="Signatures", values_to="Exposures") %>%
-
-        ggplot() +
-        geom_bar(aes(x=PatientID, y=Exposures, fill=Signatures),
-                position="stack", stat="identity") +
-        theme(axis.text.x=element_text(angle=90,hjust=1),
-                panel.background=element_blank(),
-                axis.line=element_line(colour="black"))
-
-    plt_all = patchwork::wrap_plots(plot_exposure, plot_signatures, ncol=2) + patchwork::plot_annotation(title = "$meta.id")
-    ggplot2::ggsave(plot = plt_all, filename = paste0("$prefix", "_plot_all.pdf"), width = 210, height = 297, units="mm", dpi = 200)
-    saveRDS(object = plt_all, file = paste0("$prefix", "_plot_all.rds"))
-
-    # version export
-    f <- file("versions.yml","w")
-    SparseSignatures_version <- sessionInfo()\$otherPkgs\$SparseSignatures\$Version
-    dplyr_version <- sessionInfo()\$otherPkgs\$dplyr\$Version
-    ggplot2_version <- sessionInfo()\$otherPkgs\$ggplot2\$Version
-    patchwork_version <- sessionInfo()\$otherPkgs\$patchwork\$Version
-    writeLines(paste0('"', "$task.process", '"', ":"), f)
-    writeLines(paste("    SparseSignatures:", SparseSignatures_version), f)
-    writeLines(paste("    dplyr:", dplyr_version), f)
-    writeLines(paste("    ggplot2:", ggplot2_version), f)
-    writeLines(paste("    patchwork:", patchwork_version), f)
-    close(f)
-
-    """
+    template "main_script.R"
 }

modules/local/annotate_driver/main.nf

@@ -1,8 +1,9 @@
-
 process ANNOTATE_DRIVER {
     tag "$meta.id"
     label "process_single"
-    container = 'docker://lvaleriani/cnaqc:version1.0'
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://lvaleriani/cnaqc:version1.0' :
+        'docker.io/lvaleriani/cnaqc:version1.0' }"
 
     input:
     tuple val(meta), path(rds), path(driver_list)
@@ -11,7 +12,6 @@ process ANNOTATE_DRIVER {
     tuple val(meta), path("*.rds"),     emit: rds
     path "versions.yml",                emit: versions
 
-
     script:
     def args    =   task.ext.args   ?: ''
     def prefix  =   task.ext.prefix ?: "${meta.id}"

modules/local/cna2CNAqc/main.nf

@@ -1,7 +1,9 @@
 process CNA_PROCESSING {
     tag "$meta.id"
     label "process_single"
-    container "docker://lvaleriani/cnaqc:version1.0"
+    container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
+        'docker://lvaleriani/cnaqc:version1.0' :
+        'docker.io/lvaleriani/cnaqc:version1.0' }"
 
     input:
     tuple val(meta), path(cna_segs), path(cna_extra)
@@ -14,33 +16,5 @@ process CNA_PROCESSING {
     task.ext.when == null || task.ext.when
 
     script:
-    def args    = task.ext.args     ?:  ''
-    def prefix  = task.ext.prefix   ?:  "${meta.id}"
-
-    """
-    #!/usr/bin/env Rscript
-    source(paste0("$moduleDir", '/parser_CNA.R'))
-
-    if ("$meta.cna_caller" == 'sequenza'){
-    CNA = parse_Sequenza(segments = "$cna_segs", extra = "$cna_extra")
-
-    } else if ("$meta.cna_caller" == 'ASCAT'){
-    CNA = parse_ASCAT(segments = "$cna_segs", extra = "$cna_extra")
-
-    } else {
-    stop('Copy Number Caller not supported.')
-    }
-
-    saveRDS(object = CNA, file = paste0("$prefix", "_cna.rds"))
-
-    # version export
-    f <- file("versions.yml","w")
-    readr_version <- sessionInfo()\$otherPkgs\$readr\$Version
-    dplyr_version <- sessionInfo()\$otherPkgs\$dplyr\$Version
-    writeLines(paste0('"', "$task.process", '"', ":"), f)
-    writeLines(paste("    readr:", readr_version), f)
-    writeLines(paste("    dplyr:", dplyr_version), f)
-    close(f)
-
-    """
+    template "main_script.R"
 }