# Outputs

All paths below are relative to the base directory outDir as described in the run instructions.

outDir
├── bams
├── cohort_level
├── germline
└── somatic

# BAM Files

The bams folder contains the final aligned and post-processed BAM files along with index files. It also contains FASTQ QC and basic BAM file QC.

outDir/bams/
├── DU874145-N
│   ├── DU874145-N.bam
│   ├── DU874145-N.bam.bai
│   ├── alfred
│   ├── collecthsmetrics
│   ├── fastp
│   ├── multiqc
│   └── pileup
├── DU874145-T
│   ├── DU874145-T.bam
│   └── DU874145-T.bam.bai
...

These outputs are:

  • fastp: A HTML report for each FASTQ lane pair per sample.
  • alfred: A per-sample and per-readgroup BAM file alignment metrics in text and PDF files.
  • collectshsmetrics: For exomes, per-sample hybridisation-selection metrics in the.
  • pileup: Per tumor-normal-pair, the Conpair-generated SNP pileup files.
  • multiqc: A summary report of FASTQ/BAM QC metrics from Picard, fastp and other tools.

# Somatic data

The result of the somatic analyses is output in summarized forms in the somatic folder:

outDir/somatic
├── DU874145-T__DU874145-N
│   ├── combined_mutations
│   ├── combined_svs
│   ├── conpair
│   ├── delly
│   ├── facets
│   ├── lohhla
│   ├── manta
│   ├── meta_data
│   ├── multiqc
│   ├── mutect2
│   ├── neoantigen
│   └── strelka2
└── DU874146-T__DU874146-N
    ├── combined_mutations
    ├── combined_svs
    ├── delly
    ├── facets
    ├── lohhla
    ├── manta
    ├── meta_data
    ├── multiqc
    ├── mutect2
    ├── neoantigen
    └── strelka2

These outputs are:

  • combined_mutatations: unfiltered and final filtered maf per tumor-normal pair.
    • *.somatic.unfiltered.maf: Unfiltered mutations generated in the SomaticAnnotateMaf.
    • *.somatic.final.maf: Filtered mutations from MuTect2 and Strelka2, annotated with mutational effects, neoantigen predictions, and zygosity, as described elsewhere.
    • intermidiate_files/*: 3 intermidiate vcf files contains all mutations before any filter after mutect and strelka, mutations after filter-vcf.py, and mutations after bcftools filter by FILTER=PASS.
  • combined_svs: Combined Delly and Manta SV calls.
  • conpair: Per tumor-normal-pair, the Conpair-generated concordance and contamination files.
  • delly: Delly output.
  • facets: Individual copy-number profiles from FACETS, per tumor-normal pair.
  • lohhla: LOHHLA output.
  • manta: Manta output.
  • meta_data: Summarized meta_data file which includes the following results:
    • Purity and Ploidy
    • WGS Status
    • MSI information including MSI_Total_Sites, MSI_Somatic_Sites, MSIscore
    • Number of Mutations
    • All 60 Mutational Signatures
    • HLA genotyping
    • TMB
  • multiqc: A summary report of tumor/normal pair QC metrics from Conpair and Facets.
  • mutect2: Manta output.
  • neoantigens: Neoantigen predictions from NetMHCpan per sample.
  • strelka2: Manta output.

Be aware

  • LOHHLA is temporarily disabled due to a bug need future investigation. It will be enabled again in the future release.

# Germline data

The result of the germline analyses is output in the germline folder:

outDir/germline/
├── DU874145-N
│   ├── combined_mutations
│   ├── combined_svs
│   ├── delly
│   ├── haplotypecaller
│   ├── manta
│   └── strelka2
└── DU874146-N
    ├── combined_mutations
    ├── combined_svs
    ├── delly
    ├── haplotypecaller
    ├── manta
    └── strelka2

These outputs are:

  • combined_mutatations: unfiltered and final filtered maf per tumor-normal pair.
    • *.germline.unfiltered.maf: Unfiltered mutations generated in the GermlineAnnotateMaf.
    • *.germline.final.maf: Filtered mutations from HaplotypeCaller and Strelka2, annotated with mutational effects and zygosity, as described elsewhere.
    • intermidiate_files/*: 3 intermidiate vcf files contains all mutations before any filter after mutect and strelka, mutations after bcftools filter by FILTER=PASS, and gnomAD filter.
  • combined_svs: Combined Delly and Manta SV calls.
  • delly: Delly output.
  • manta: Manta output.
  • strelka2: Manta output.

# Cohort Level Outputs

When run with the flag --aggregate, the pipeline will output aggregate all samples together for each processes and output as a single file for each processes. The files are:

outDir/cohort_level/
├── default_cohort
│   ├── alignment_qc.txt
│   ├── cna_armlevel.txt
│   ├── cna_facets_run_info.txt
│   ├── cna_genelevel.txt
│   ├── cna_hisens_run_segmentation.seg
│   ├── cna_purity_run_segmentation.seg
│   ├── concordance_qc.txt
│   ├── contamination_qc.txt
│   ├── DNA.IntegerCPN_CI.txt
│   ├── HLAlossPrediction_CI.txt
│   ├── multiqc_report.html
│   ├── multiqc_data.zip
│   ├── mut_germline.maf
│   ├── mut_somatic.maf
│   ├── mut_somatic_neoantigens.txt
│   ├── sample_data.txt
│   ├── sv_germline.vcf.gz
│   ├── sv_germline.vcf.gz.tbi
│   ├── sv_somatic.vcf.gz
│   └── sv_somatic.vcf.gz.tbi
├── cohort2
│   ├── alignment_qc.txt
│   ├── cna_armlevel.txt
│   ├── cna_facets_run_info.txt
│   ├── cna_genelevel.txt
...

These outputs are just naively concatenated together from per sample output files (duplicated header are removed). A combined MultiQC summary report is produced containing QC metrics for all samples and tumor/normal pairs in the cohort.