main.nf

#!/usr/bin/env nextflow
/*
========================================================================================
                         nf-core/metatdenovo
========================================================================================
 nf-core/metatdenovo Analysis Pipeline.
 #### Homepage / Documentation
 https://github.com/nf-core/metatdenovo
----------------------------------------------------------------------------------------
*/

def helpMessage() {
    log.info nfcoreHeader()
    log.info"""

    Usage:

    The typical command for running the pipeline is as follows:

    nextflow run nf-core/metatdenovo --reads '*_R{1,2}.fastq.gz' -profile docker

    Mandatory arguments:
      --reads                       Path to input data (must be surrounded with quotes)
      -profile                      Configuration profile to use. Can use multiple (comma separated)
                                    Available: conda, docker, singularity, awsbatch, test and more.

    Options:
      --megahit			    Run assembly with Megahit.
      --trinity			    Run assembly with Trinity. This is done in a very unsophisticated way, see usage.md.
      --singleEnd                   Specifies that the input is single end reads. Currently NOT SUPPORTED.

    Other options:
      --outdir                      The output directory where the results will be saved. (Default: 'results/')
      --email                       Set this parameter to your e-mail address to get a summary e-mail with details of the run sent to you when the workflow exits
      --email_on_fail               Same as --email, except only send mail if the workflow is not successful
      --maxMultiqcEmailFileSize     Theshold size for MultiQC report to be attached in notification email. If file generated by pipeline exceeds the threshold, it will not be attached (Default: 25MB)
      -name                         Name for the pipeline run. If not specified, Nextflow will automatically generate a random mnemonic.

    AWSBatch options:
      --awsqueue                    The AWSBatch JobQueue that needs to be set when running on AWSBatch
      --awsregion                   The AWS Region for your AWS Batch job to run on
    """.stripIndent()
}

// Show help message
if (params.help) {
    helpMessage()
    exit 0
}

/*
 * SET UP CONFIGURATION VARIABLES
 */

params.outdir = 'results/'
params.megahit = false // Run, or not, megahit assembly
params.trinity = false // Run, or not, trinity assembly

// Has the run name been specified by the user?
//  this has the bonus effect of catching both -name and --name
custom_runName = params.name
if (!(workflow.runName ==~ /[a-z]+_[a-z]+/)) {
  custom_runName = workflow.runName
}

if ( workflow.profile == 'awsbatch') {
  // AWSBatch sanity checking
  if (!params.awsqueue || !params.awsregion) exit 1, "Specify correct --awsqueue and --awsregion parameters on AWSBatch!"
  // Check outdir paths to be S3 buckets if running on AWSBatch
  // related: https://github.com/nextflow-io/nextflow/issues/813
  if (!params.outdir.startsWith('s3:')) exit 1, "Outdir not on S3 - specify S3 Bucket to run on AWSBatch!"
  // Prevent trace files to be stored on S3 since S3 does not support rolling files.
  if (workflow.tracedir.startsWith('s3:')) exit 1, "Specify a local tracedir or run without trace! S3 cannot be used for tracefiles."
}

// Stage config files
ch_multiqc_config = file(params.multiqc_config, checkIfExists: true)
ch_output_docs = file("$baseDir/docs/output.md", checkIfExists: true)

/*
 * Create a channel for input read files
 */
if (params.readPaths) {
    if (params.singleEnd) {
        Channel
            .from(params.readPaths)
            .map { row -> [ row[0], [ file(row[1][0], checkIfExists: true) ] ] }
            .ifEmpty { exit 1, "params.readPaths was empty - no input files supplied" }
            .into { read_files_fastqc; read_files_trimming }
    } else {
        Channel
            .from(params.readPaths)
            .map { row -> [ row[0], [ file(row[1][0], checkIfExists: true), file(row[1][1], checkIfExists: true) ] ] }
            .ifEmpty { exit 1, "params.readPaths was empty - no input files supplied" }
            .into { read_files_fastqc; read_files_trimming }
    }
} else {
    Channel
        .fromFilePairs( params.reads, size: params.singleEnd ? 1 : 2 )
        .ifEmpty { exit 1, "Cannot find any reads matching: ${params.reads}\nNB: Path needs to be enclosed in quotes!\nIf this is single-end data, please specify --singleEnd on the command line." }
        .into { read_files_fastqc; read_files_trimming }
}

// Header log info
log.info nfcoreHeader()
def summary = [:]
if (workflow.revision) summary['Pipeline Release'] = workflow.revision
summary['Run Name']         = custom_runName ?: workflow.runName
// TODO nf-core: Report custom parameters here
summary['Reads']            = params.reads
summary['Data Type']        = params.singleEnd ? 'Single-End' : 'Paired-End'
summary['Max Resources']    = "$params.max_memory memory, $params.max_cpus cpus, $params.max_time time per job"
if (workflow.containerEngine) summary['Container'] = "$workflow.containerEngine - $workflow.container"
summary['Output dir']       = params.outdir
summary['Launch dir']       = workflow.launchDir
summary['Working dir']      = workflow.workDir
summary['Script dir']       = workflow.projectDir
summary['User']             = workflow.userName
if (workflow.profile == 'awsbatch') {
  summary['AWS Region']     = params.awsregion
  summary['AWS Queue']      = params.awsqueue
}
summary['Config Profile'] = workflow.profile
if (params.config_profile_description) summary['Config Description'] = params.config_profile_description
if (params.config_profile_contact)     summary['Config Contact']     = params.config_profile_contact
if (params.config_profile_url)         summary['Config URL']         = params.config_profile_url
if (params.email || params.email_on_fail) {
  summary['E-mail Address']    = params.email
  summary['E-mail on failure'] = params.email_on_fail
  summary['MultiQC maxsize']   = params.maxMultiqcEmailFileSize
}
log.info summary.collect { k,v -> "${k.padRight(18)}: $v" }.join("\n")
log.info "-\033[2m--------------------------------------------------\033[0m-"

// Check the hostnames against configured profiles
checkHostname()

def create_workflow_summary(summary) {
    def yaml_file = workDir.resolve('workflow_summary_mqc.yaml')
    yaml_file.text  = """
    id: 'nf-core-metatdenovo-summary'
    description: " - this information is collected when the pipeline is started."
    section_name: 'nf-core/metatdenovo Workflow Summary'
    section_href: 'https://github.com/nf-core/metatdenovo'
    plot_type: 'html'
    data: |
        <dl class=\"dl-horizontal\">
${summary.collect { k,v -> "            <dt>$k</dt><dd><samp>${v ?: '<span style=\"color:#999999;\">N/A</a>'}</samp></dd>" }.join("\n")}
        </dl>
    """.stripIndent()

   return yaml_file
}

/*
 * Parse software version numbers
 */
process get_software_versions {
    publishDir "${params.outdir}/pipeline_info", mode: 'copy',
        saveAs: { filename ->
            if (filename.indexOf(".csv") > 0) filename
            else null
        }

    output:
    file 'software_versions_mqc.yaml' into software_versions_yaml
    file "software_versions.csv"

    script:
    // TODO nf-core: Get all tools to print their version number here
    """
    echo $workflow.manifest.version > v_pipeline.txt
    echo $workflow.nextflow.version > v_nextflow.txt
    fastqc --version > v_fastqc.txt
    multiqc --version > v_multiqc.txt
    cutadapt --version > v_cutadapt.txt
    trim_galore --version | grep version > v_trim_galore.txt
    megahit --version > v_megahit.txt
    grep "my \\+\\\$VERSION" \$(which Trinity) |grep -v "#"|sed 's/.*"\\(.*\\)"; */\\1/' > v_trinity.txt
    scrape_software_versions.py &> software_versions_mqc.yaml
    """
}

/*
 * STEP 1 - FastQC
 */
process fastqc {
    tag "$name"
    label 'process_medium'
    publishDir "${params.outdir}/fastqc", mode: 'copy',
        saveAs: { filename -> filename.indexOf(".zip") > 0 ? "zips/$filename" : "$filename" }

    input:
    set val(name), file(reads) from read_files_fastqc

    output:
    file "*_fastqc.{zip,html}" into fastqc_results

    script:
    """
    fastqc --quiet --threads $task.cpus $reads
    """
}

/*
 * STEP 2 - MultiQC
 */
process multiqc {
    publishDir "${params.outdir}/MultiQC", mode: 'copy'

    input:
    file multiqc_config from ch_multiqc_config
    // TODO nf-core: Add in log files from your new processes for MultiQC to find!
    file ('fastqc/*') from fastqc_results.collect().ifEmpty([])
    file ('software_versions/*') from software_versions_yaml.collect()
    file workflow_summary from create_workflow_summary(summary)

    output:
    file "*multiqc_report.html" into multiqc_report
    file "*_data"
    file "multiqc_plots"

    script:
    rtitle = custom_runName ? "--title \"$custom_runName\"" : ''
    rfilename = custom_runName ? "--filename " + custom_runName.replaceAll('\\W','_').replaceAll('_+','_') + "_multiqc_report" : ''
    // TODO nf-core: Specify which MultiQC modules to use with -m for a faster run time
    """
    multiqc -f $rtitle $rfilename --config $multiqc_config .
    """
}

/*
 * STEP 3 - Output Description HTML
 */
process output_documentation {
    publishDir "${params.outdir}/pipeline_info", mode: 'copy'

    input:
    file output_docs from ch_output_docs

    output:
    file "results_description.html"

    script:
    """
    markdown_to_html.r $output_docs results_description.html
    """
}

/*
 * STEP 4 - Trimming
 */
process trim_galore {
  cpus 1
  time params.max_time

  publishDir("${params.outdir}/trimming_logs/", mode: "copy", pattern: "*.trim_galore.log")

  input:
  tuple name, file(reads) from read_files_trimming

  output:
  //tuple val(name), file("*_1.fq.gz") into (
  file("*_1.fq.gz") into (
    trimmed_fwdreads_megahit, trimmed_fwdreads_trinity
  )
  //tuple val(name), file("*_2.fq.gz") into (
  file("*_2.fq.gz") into (
    trimmed_revreads_megahit, trimmed_revreads_trinity
  )
  tuple val(name), file("*.trim_galore.log") into trimming_logs
  // TODO: Check how to best get this into fastqc/multiqc
  //file "*_fastqc.{zip,html}" into trimgalore_fastqc_reports

  """
  trim_galore --paired --fastqc --gzip --quality 20 $reads 2>&1 > ${name}.trim_galore.log
  """
}

/*
 * STEP 5a - Megahit assembly
 */
process megahit {
  cpus params.max_cpus

  publishDir("${params.outdir}/megahit", mode: "copy")

  when:
  params.megahit

  input:
  file(fwdreads) from trimmed_fwdreads_megahit.collect()
  file(revreads) from trimmed_revreads_megahit.collect()

  output:
  file "megahit.final.contigs.fna.gz"
  file "megahit.log"
  file "megahit.tar.gz"

  """
  megahit -t ${task.cpus} -1 ${fwdreads.sort().join(',')} -2 ${revreads.sort().join(',')} > megahit.log 2>&1
  cp megahit_out/final.contigs.fa megahit.final.contigs.fna
  pigz -p ${task.cpus} megahit.final.contigs.fna
  tar cfz megahit.tar.gz megahit_out/
  """
}

/*
 * STEP 5b - Trinity assembly
 */
process trinity {
  cpus params.max_cpus

  publishDir("${params.outdir}/trinity", mode: "copy")

  when:
  params.trinity

  input:
  file(fwdreads) from trimmed_fwdreads_trinity.collect()
  file(revreads) from trimmed_revreads_trinity.collect()

  output:
  file "trinity.final.contigs.fna.gz"
  file "trinity.log"
  file "trinity.tar.gz"

  """
  unpigz -c -p ${task.cpus} ${fwdreads.sort()} > fwdreads.fastq
  unpigz -c -p ${task.cpus} ${revreads.sort()} > revreads.fastq
  Trinity --seqType fq --max_memory \$(echo ${task.memory}|sed 's/ *GB/G/') --left fwdreads.fastq --right revreads.fastq > trinity.log 2>&1
  cp trinity_out_dir/Trinity.fasta trinity.final.contigs.fna
  pigz -p ${task.cpus} trinity.final.contigs.fna
  tar cfz trinity.tar.gz trinity_out_dir/
  """
}


/*
 * Completion e-mail notification
 */
workflow.onComplete {

    // Set up the e-mail variables
    def subject = "[nf-core/metatdenovo] Successful: $workflow.runName"
    if (!workflow.success) {
      subject = "[nf-core/metatdenovo] FAILED: $workflow.runName"
    }
    def email_fields = [:]
    email_fields['version'] = workflow.manifest.version
    email_fields['runName'] = custom_runName ?: workflow.runName
    email_fields['success'] = workflow.success
    email_fields['dateComplete'] = workflow.complete
    email_fields['duration'] = workflow.duration
    email_fields['exitStatus'] = workflow.exitStatus
    email_fields['errorMessage'] = (workflow.errorMessage ?: 'None')
    email_fields['errorReport'] = (workflow.errorReport ?: 'None')
    email_fields['commandLine'] = workflow.commandLine
    email_fields['projectDir'] = workflow.projectDir
    email_fields['summary'] = summary
    email_fields['summary']['Date Started'] = workflow.start
    email_fields['summary']['Date Completed'] = workflow.complete
    email_fields['summary']['Pipeline script file path'] = workflow.scriptFile
    email_fields['summary']['Pipeline script hash ID'] = workflow.scriptId
    if (workflow.repository) email_fields['summary']['Pipeline repository Git URL'] = workflow.repository
    if (workflow.commitId) email_fields['summary']['Pipeline repository Git Commit'] = workflow.commitId
    if (workflow.revision) email_fields['summary']['Pipeline Git branch/tag'] = workflow.revision
    if (workflow.container) email_fields['summary']['Docker image'] = workflow.container
    email_fields['summary']['Nextflow Version'] = workflow.nextflow.version
    email_fields['summary']['Nextflow Build'] = workflow.nextflow.build
    email_fields['summary']['Nextflow Compile Timestamp'] = workflow.nextflow.timestamp

    // TODO nf-core: If not using MultiQC, strip out this code (including params.maxMultiqcEmailFileSize)
    // On success try attach the multiqc report
    def mqc_report = null
    try {
        if (workflow.success) {
            mqc_report = multiqc_report.getVal()
            if (mqc_report.getClass() == ArrayList) {
                log.warn "[nf-core/metatdenovo] Found multiple reports from process 'multiqc', will use only one"
                mqc_report = mqc_report[0]
            }
        }
    } catch (all) {
        log.warn "[nf-core/metatdenovo] Could not attach MultiQC report to summary email"
    }

    // Check if we are only sending emails on failure
    email_address = params.email
    if (!params.email && params.email_on_fail && !workflow.success) {
        email_address = params.email_on_fail
    }

    // Render the TXT template
    def engine = new groovy.text.GStringTemplateEngine()
    def tf = new File("$baseDir/assets/email_template.txt")
    def txt_template = engine.createTemplate(tf).make(email_fields)
    def email_txt = txt_template.toString()

    // Render the HTML template
    def hf = new File("$baseDir/assets/email_template.html")
    def html_template = engine.createTemplate(hf).make(email_fields)
    def email_html = html_template.toString()

    // Render the sendmail template
    def smail_fields = [ email: email_address, subject: subject, email_txt: email_txt, email_html: email_html, baseDir: "$baseDir", mqcFile: mqc_report, mqcMaxSize: params.maxMultiqcEmailFileSize.toBytes() ]
    def sf = new File("$baseDir/assets/sendmail_template.txt")
    def sendmail_template = engine.createTemplate(sf).make(smail_fields)
    def sendmail_html = sendmail_template.toString()

    // Send the HTML e-mail
    if (email_address) {
        try {
          if ( params.plaintext_email ){ throw GroovyException('Send plaintext e-mail, not HTML') }
          // Try to send HTML e-mail using sendmail
          [ 'sendmail', '-t' ].execute() << sendmail_html
          log.info "[nf-core/metatdenovo] Sent summary e-mail to $email_address (sendmail)"
        } catch (all) {
          // Catch failures and try with plaintext
          [ 'mail', '-s', subject, email_address ].execute() << email_txt
          log.info "[nf-core/metatdenovo] Sent summary e-mail to $email_address (mail)"
        }
    }

    // Write summary e-mail HTML to a file
    def output_d = new File( "${params.outdir}/pipeline_info/" )
    if (!output_d.exists()) {
      output_d.mkdirs()
    }
    def output_hf = new File( output_d, "pipeline_report.html" )
    output_hf.withWriter { w -> w << email_html }
    def output_tf = new File( output_d, "pipeline_report.txt" )
    output_tf.withWriter { w -> w << email_txt }

    c_reset = params.monochrome_logs ? '' : "\033[0m";
    c_purple = params.monochrome_logs ? '' : "\033[0;35m";
    c_green = params.monochrome_logs ? '' : "\033[0;32m";
    c_red = params.monochrome_logs ? '' : "\033[0;31m";

    if (workflow.stats.ignoredCount > 0 && workflow.success) {
      log.info "${c_purple}Warning, pipeline completed, but with errored process(es) ${c_reset}"
      log.info "${c_red}Number of ignored errored process(es) : ${workflow.stats.ignoredCount} ${c_reset}"
      log.info "${c_green}Number of successfully ran process(es) : ${workflow.stats.succeedCount} ${c_reset}"
    }

    if (workflow.success) {
        log.info "${c_purple}[nf-core/metatdenovo]${c_green} Pipeline completed successfully${c_reset}"
    } else {
        checkHostname()
        log.info "${c_purple}[nf-core/metatdenovo]${c_red} Pipeline completed with errors${c_reset}"
    }

}


def nfcoreHeader(){
    // Log colors ANSI codes
    c_reset = params.monochrome_logs ? '' : "\033[0m";
    c_dim = params.monochrome_logs ? '' : "\033[2m";
    c_black = params.monochrome_logs ? '' : "\033[0;30m";
    c_green = params.monochrome_logs ? '' : "\033[0;32m";
    c_yellow = params.monochrome_logs ? '' : "\033[0;33m";
    c_blue = params.monochrome_logs ? '' : "\033[0;34m";
    c_purple = params.monochrome_logs ? '' : "\033[0;35m";
    c_cyan = params.monochrome_logs ? '' : "\033[0;36m";
    c_white = params.monochrome_logs ? '' : "\033[0;37m";

    return """    -${c_dim}--------------------------------------------------${c_reset}-
                                            ${c_green},--.${c_black}/${c_green},-.${c_reset}
    ${c_blue}        ___     __   __   __   ___     ${c_green}/,-._.--~\'${c_reset}
    ${c_blue}  |\\ | |__  __ /  ` /  \\ |__) |__         ${c_yellow}}  {${c_reset}
    ${c_blue}  | \\| |       \\__, \\__/ |  \\ |___     ${c_green}\\`-._,-`-,${c_reset}
                                            ${c_green}`._,._,\'${c_reset}
    ${c_purple}  nf-core/metatdenovo v${workflow.manifest.version}${c_reset}
    -${c_dim}--------------------------------------------------${c_reset}-
    """.stripIndent()
}

def checkHostname(){
    def c_reset = params.monochrome_logs ? '' : "\033[0m"
    def c_white = params.monochrome_logs ? '' : "\033[0;37m"
    def c_red = params.monochrome_logs ? '' : "\033[1;91m"
    def c_yellow_bold = params.monochrome_logs ? '' : "\033[1;93m"
    if (params.hostnames) {
        def hostname = "hostname".execute().text.trim()
        params.hostnames.each { prof, hnames ->
            hnames.each { hname ->
                if (hostname.contains(hname) && !workflow.profile.contains(prof)) {
                    log.error "====================================================\n" +
                            "  ${c_red}WARNING!${c_reset} You are running with `-profile $workflow.profile`\n" +
                            "  but your machine hostname is ${c_white}'$hostname'${c_reset}\n" +
                            "  ${c_yellow_bold}It's highly recommended that you use `-profile $prof${c_reset}`\n" +
                            "============================================================"
                }
            }
        }
    }
}