HLA typing - mapping of RNAseq fastqs - avoiding RazerS ressource dependencies

[PeterBailey]

The following code below was modified in main.nf.

Simply removed the unzip process and replaced with modified yara mapper script.

The code was run as follows:

nextflow run ~/nf_scripts/hlaTypingMaster -profile docker --reads '*_R{1,2}.fastq.gz' --seqtype 'rna' --outdir $PWD/SCC_hlatyping --index data/indices/yara/hla _reference_rna

Received error:

Jul-30 14:01:33.586 [Task monitor] ERROR nextflow.processor.TaskProcessor - Error executing process > 'map_fastq_to_hla (5)'

Caused by:
  Process `map_fastq_to_hla (5)` terminated with an error exit status (1)

Command executed:

  yara_mapper -e 3 -t 16 /home/pbailey/nf_scripts/hlaTypingMaster/data/indices/yara/hla_reference_rna MET2_R1.fastq.gz MET2_R2.fastq.gz -o output.bam
  samtools view -h -F 4 -f 0x40 -b1 output.bam > mapped_1.bam
  samtools view -h -F 4 -f 0x80 -b1 output.bam > mapped_2.bam

Command exit status:
  1

Command output:
  (empty)

Command error:
  Open failed on file /home/pbailey/nf_scripts/hlaTypingMaster/data/indices/yara/hla_reference_rna.txt.size: "No such file or directory"
  yara_mapper: Error while opening reference file.

if( params.bam ) log.info "BAM file format detected. Initiate remapping to HLA alleles with yara mapper."

/*
 * Preparation - Unpack files if packed.
 * 
 * OptiType cannot handle *.gz archives as input files, 
 * So we have to unpack first, if this is the case. 
 */
if ( !params.bam  ) { // FASTQ files processing
    process map_fastq_to_hla {
        input:
        set val(pattern), file(reads) from input_data

        output:
        set val(pattern), "mapped_{1,2}.bam" into raw_reads

        script:
        if (params.singleEnd)
        """
        yara_mapper -e 3 -t ${params.max_cpus} -f bam ${workflow.projectDir}/${params.index} ${reads[0]}  > output_1.bam
        samtools view -h -F 4 -b1 output_1.bam > mapped_1.bam
        """
        else
        """
        yara_mapper -e 3 -t ${params.max_cpus} ${workflow.projectDir}/${params.index} ${reads[0]} ${reads[1]} -o output.bam
        samtools view -h -F 4 -f 0x40 -b1 output.bam > mapped_1.bam
        samtools view -h -F 4 -f 0x80 -b1 output.bam > mapped_2.bam
        """
    }
} else { // BAM files processing

    /*
     * Preparation - Remapping of reads against HLA reference and filtering these
     *
     * In case the user provides BAM files, a remapping step
     * is then done against the HLA reference sequence.
     */
    process remap_to_hla {
        
        input:
        set val(pattern), file(bams) from input_data

        output:
        set val(pattern), "mapped_{1,2}.bam" into raw_reads

        script:
        if (params.singleEnd)
        """
        samtools bam2fq $bams > output_1.fastq
        yara_mapper -e 3 -t ${params.max_cpus} -f bam ${workflow.projectDir}/${params.index} output_1.fastq > output_1.bam
        samtools view -h -F 4 -b1 output_1.bam > mapped_1.bam
        """
        else
        """
        samtools view -h -f 0x40 $bams > output_1.bam
        samtools view -h -f 0x80 $bams > output_2.bam
        samtools bam2fq output_1.bam > output_1.fastq
        samtools bam2fq output_2.bam > output_2.fastq
        yara_mapper -e 3 -t ${params.max_cpus} -f bam ${workflow.projectDir}/${params.index} output_1.fastq output_2.fastq > output.bam
        samtools view -h -F 4 -f 0x40 -b1 output.bam > mapped_1.bam
        samtools view -h -F 4 -f 0x80 -b1 output.bam > mapped_2.bam
        """

    }

}

[sven1103]

thanks @PeterBailey for reporting this. We will implement a pre-mapping step with yara in the next minor release of the HLA typing pipleline!

[sven1103]

solved in #15