omicverse.alignment.dada2_pipeline

Contents

omicverse.alignment.dada2_pipeline#

omicverse.alignment.dada2_pipeline(samples, workdir=None, *, db_fasta=None, trunc_len=(240, 160), max_ee=(2.0, 2.0), trunc_q=2, min_overlap=12, max_mismatch=0, chimera_method='consensus', nbases=100000000, sintax_cutoff=0.8, sintax_strand='both', threads=4, sample_metadata=None, overwrite=False)[source]#

Run pydada2 end-to-end and return an AnnData.

Parameters:

  • samples (Sequence[Tuple[str, str, Optional[str]]]) – [(sample, fq1, fq2), ...]. fq2 may be None for single-end.

  • workdir (Optional[str] (default: None)) – Required. Absolute path for intermediates. No $HOME fallback.

  • db_fasta (Optional[str] (default: None)) – Optional path to a SINTAX-formatted reference FASTA. When supplied, ASV taxonomy is assigned by piping the ASV FASTA through omicverse.alignment.vsearch.sintax() and merging the 7-rank call into adata.var.

  • trunc_len (Union[int, Tuple[int, int]] (default: (240, 160))) – (fwd, rev) truncation lengths in bp. V4 default (240, 160).

  • max_ee (Union[float, Tuple[float, float]] (default: (2.0, 2.0))) – (fwd, rev) expected-error thresholds.

  • chimera_method (str (default: 'consensus')) – Passed to pydada2.remove_bimera_denovo.

  • threads (int (default: 4)) – Used only by the vsearch SINTAX pass at the end.

Contents