nextflow.enable.dsl = 2

include { CALL_WF } from './workflows/call_wf.nf'

include { VALIDATE_FASTQS_WF } from './workflows/validate_fastqs_wf.nf'

include { MAP_WF } from './workflows/map_wf.nf'

include { MERGE_WF } from './workflows/merge_wf.nf'

include { MINOR_VARIANT_ANALYSIS_WF } from './workflows/minor_variant_analysis_wf.nf'

include { QUALITY_CHECK_WF } from './workflows/quality_check_wf.nf'

include { REPORTS_WF } from './workflows/reports_wf.nf'

workflow {

if (params.only_validate_fastqs) {

VALIDATE_FASTQS_WF(params.input_samplesheet)

} else {

validated_reads_ch = VALIDATE_FASTQS_WF( params.input_samplesheet )

QUALITY_CHECK_WF( validated_reads_ch )

MAP_WF( validated_reads_ch )

CALL_WF( MAP_WF.out.sorted_reads_ch )

MINOR_VARIANT_ANALYSIS_WF(CALL_WF.out.reformatted_lofreq_vcfs_tuple_ch)

//---------------------------------------------------------------------------------

// Filter the approved samples

//---------------------------------------------------------------------------------

//NOTE: Read the approved_samples tsv file and isolate the names of the approved samples

approved_samples_minor_variants_ch = MINOR_VARIANT_ANALYSIS_WF.out.approved_samples_ch

.splitCsv(header: false, skip: 1, sep: '\t' )

.map { row -> [ row.first() ] }

.collect()

.dump(tag:'approved_samples_minor_variants_ch')

/* .view {"\n\n XBS-NF-LOG approved_samples_minor_variants_ch : $it \n\n"} */

//NOTE: Reshape the flattened output of gvch_ch into the tuples of [sampleName, gvcf, gvcf.tbi]

collated_gvcfs_ch = CALL_WF.out.gvcf_ch

.flatten()

.collate(3)

.dump(tag:'collated_gvcfs_ch')

/* .view {"\n\n XBS-NF-LOG collated_gvcfs_ch : $it \n\n"} */

//.collectFile(name: "$params.outdir/collated_gvcfs_ch.txt")

//FIXME: Refactor this to emit two different files and use only the approved samples

//NOTE: Use the stats file for the entire cohort (from CALL_WF)

// and filter out the samples which pass all thresholds

approved_call_wf_samples_ch = CALL_WF.out.cohort_stats_tsv

.splitCsv(header: false, skip: 1, sep: '\t' )

.map { row -> [

row.first(), // SAMPLE

row.last().toInteger() // ALL_THRESHOLDS_MET

]

}

.filter { it[1] == 1} // Filter out samples which meet all the thresholds

.map { [ it[0] ] }

approved_call_wf_samples_ch

.collect()

.dump(tag:'approved_call_wf_samples_ch.collect()')

/* .view {"\n\n XBS-NF-LOG approved_call_wf_samples_ch.collect() : $it \n\n"} */

//NOTE: Join the approved samples from MINOR_VARIANT_ANALYSIS_WF and CALL_WF

fully_approved_samples_ch = approved_samples_minor_variants_ch

.join(approved_call_wf_samples_ch)

.flatten()

.dump(tag:'fully_approved_samples_ch')

/* .view {"\n\n XBS-NF-LOG fully_approved_samples_ch : $it \n\n"} */

//.collect()

//.collectFile(name: "$params.outdir/approved_samples_ch.txt")

//NOTE: Join the fully approved samples with the gvcf channel to select files for MERGE_WF

selected_gvcfs_ch = collated_gvcfs_ch

.join(fully_approved_samples_ch)

.flatten()

.filter { it.class == sun.nio.fs.UnixPath }

.collect()

.dump(tag:'selected_gvcfs_ch')

/* .view {"\n\n XBS-NF-LOG selected_gvcfs_ch : $it \n\n"} */

//.collectFile(name: "$params.outdir/selected_gvcfs_ch")

//---------------------------------------------------------------------------------

MERGE_WF(selected_gvcfs_ch, CALL_WF.out.reformatted_lofreq_vcfs_tuple_ch)

REPORTS_WF(QUALITY_CHECK_WF.out.reports_fastqc_ch)

}

}