omicverse.alignment.amplicon_16s_pipeline

omicverse.alignment.amplicon_16s_pipeline#

omicverse.alignment.amplicon_16s_pipeline(fastq_dir=None, samples=None, workdir=None, db_fasta=None, *, primer_fwd=None, primer_rev=None, backend='vsearch', threads=4, jobs=None, merge_max_diffs=10, merge_min_overlap=16, filter_max_ee=1.0, filter_min_len=0, filter_max_len=0, derep_min_uniq=2, unoise_alpha=2.0, unoise_minsize=2, chimera_removal=True, otutab_identity=0.97, sintax_cutoff=0.8, sintax_strand='both', sample_metadata=None, overwrite=False)[source]#

Run the full 16S amplicon pipeline.

Parameters:

  • fastq_dir (Optional[str] (default: None)) – Directory containing paired Illumina FASTQs. Samples are auto-discovered by R1/R2 naming. Mutually exclusive with samples.

  • samples (Optional[Sequence[Tuple[str, str, Optional[str]]]] (default: None)) – Explicit list of (sample, fq1, fq2) tuples (fq2 may be None for single-end). Mutually exclusive with fastq_dir.

  • workdir (Optional[str] (default: None)) – Root directory for all intermediate files. No $HOME fallback.

  • db_fasta (Optional[str] (default: None)) – Path to a SINTAX-formatted 16S reference FASTA (.fa or .fa.gz). If None, taxonomy assignment is skipped.

  • primer_fwd (Optional[str] (default: None)) – PCR primer sequences. When both are provided, cutadapt() runs first; otherwise primer trimming is skipped (e.g. the mothur MiSeq SOP test dataset ships with primers already removed).

  • primer_rev (Optional[str] (default: None)) – PCR primer sequences. When both are provided, cutadapt() runs first; otherwise primer trimming is skipped (e.g. the mothur MiSeq SOP test dataset ships with primers already removed).

  • backend (str (default: 'vsearch')) – 'vsearch' (default) — UNOISE3 via vsearch. 'dada2' — pure-Python DADA2 via omicverse.alignment.dada2_pipeline() (needs pip install pydada2). 'emu' / 'qiime2' still raise NotImplementedError.

  • threads (int (default: 4)) – CPU parallelism.

  • jobs (Optional[int] (default: None)) – CPU parallelism.

  • overwrite (bool (default: False)) – If True, re-run each step regardless of existing outputs.

Returns:

Samples × ASVs matrix with taxonomy / sequence / confidence in var. Sample metadata (if provided) is merged into obs.

Return type:

anndata.AnnData

Contents