Skip to content

README.md

Latest commit

 

History

History
33 lines (19 loc) · 1.39 KB

README.md

File metadata and controls

33 lines (19 loc) · 1.39 KB
Raw

Metagenomic preprocessing

Raw paired-end metagenomic reads were processed on CSC Puhti before taxonomic and functional profiling. Scripts are in raw_reads/.

Raw read processing (taxprofiler)

Pipeline: nf-core/taxprofiler v1.1.5 (Nextflow).

Steps enabled in raw_reads/taxprofiler_runs.sh:

  1. QC — FastQC
  2. Trimming — fastp (adapters and low-quality bases)
  3. Complexity filter — BBduk
  4. Host removal — Bowtie2 against human reference T2T-CHM13v2
  5. Taxonomy — MetaPhlAn4 (merge_metaphlan_tables.py for merged profiles)

Run example: sbatch raw_reads/taxprofiler_runs.sh (edit samplesheet and config paths for your Puhti project).

More setup notes: workflow_metagenome.

Functional profiling (optional)

HUMAnN3 via Snakemake (raw_reads/Snakefile_humann.txt, submit with raw_reads/run_humann_workflow.sh) on host-depleted FASTQs (*.unmapped_1.fastq.gz). Post-processing: raw_reads/humann_modif.sh.

Downstream (this repo)

MetaPhlAn abundance tables used in the Quarto pipeline are cleaned with:

  • remove_columns_metaphlan_db_meta4_combined_reports.R — drop sample columns and write latest_metaphlan_db_meta4_combined_reports.txt

Note on PCR deduplication

Read deduplication was not included in this workflow.