bam_processing

Purpose

Filters and prepares BAM input for pileup generation.

Upstream

None (DAG root step).

Required inputs

  • bam_file (from config/input resolution)
  • optional tissue_position for in-tissue barcode extraction
  • barcode_key (for example CB)

Input interpretation

Input key Source Required Interpretation
bam_file input_details.bam_file or spaceranger_dir Yes Primary aligned BAM. Missing file aborts step execution.
tissue_position input_details.tissue_position or spaceranger_dir/spatial/* Conditional Required for Visium in-tissue barcode filtering; otherwise this stage can pass BAM through.
barcode_key input_details.barcode_key Yes BAM tag containing barcode identity used by barcode-based filtering logic.

Parameters

From steps.bam_processing:

  • nm_threshold: max mismatch tag (nM) allowed
  • mapq_threshold: minimum MAPQ

Parameter interpretation

Parameter Type Typical/default Interpretation
nm_threshold int 5 (template) Maximum allowed mismatch count per read; lower values are stricter.
mapq_threshold int 255 (template) Minimum mapping quality threshold; lower values are more permissive.

Outputs

  • in_bam: output_dir/bam_processing/IN.bam
  • in_filter_bam: output_dir/bam_processing/IN_filter.bam

Tuning notes

  • If tissue positions are provided, sinto filterbarcodes is used to keep in-tissue barcodes.
  • Otherwise, raw BAM is symlinked as IN.bam.
  • Filtered BAM is indexed for downstream steps.