BarcodeQC

At a glance

Repository: atlasxomics/barcodeqc_wf (Workflow) · atlasxomics/barcodeqc (CLI) · Display name: Barcode QC

Overview

BarcodeQC assesses barcode quality for a sequencing run from the read 2 (barcode) FASTQ — a fast early check before committing to a full preprocessing run. It serves two purposes:

1. Quality assessment — an interactive HTML report of barcode match rates and their spatial distribution, so you can spot failed lanes and low-quality runs up front.
2. Correction tables — it detects microfluidic artifacts and generates the cleaning / cross-talk correction tables (cleanTable.csv) that preprocessing consumes as its ds_table.

Running BarcodeQC before preprocessing is what lets the cleaning / cross-talk steps run inline during preprocessing rather than as a later remediation loop (see the Epigenomics overview).

Steps

1. bc_task — Matches read-2 barcodes against the selected barcode set's k-mer whitelist (merChecker), builds the per-barcode spatial table, then runs the detection / cleaning scripts (detectNclean_v1, detectNClean_smoothSlice_v2 across several extension / patch variants) to flag outlier rows/columns and produce the correction table and spatial QC plots. Reads are downsampled to outReads first.

Inputs

Parameter	Type	Default	Description
sampleName	str	—	Sample name (used to name outputs).
remoteReadTwo	LatchFile	—	Read 2 (barcode) FASTQ.
bcSet	str	bc220	Barcode set — bc220 | fg96 | bc96 | bc50.
outReads	int	10000000	Downsample target (reads to assess).
seed	int	100	Random seed for downsampling.
tissuePos_file	LatchFile	(none)	Optional tissue positions file.
output_directory	LatchDir	latch:///bcQC_output	Output location.

Outputs

Written to output_directory (default latch:///bcQC_output/), under a <sample>/ directory.

bcQC_output/
└── <sample>/
    ├── <sample>_report.html            # interactive barcode QC report
    ├── <sample>_cleanTable.csv         # cleaning / cross-talk correction table (preprocessing ds_table)
    ├── <sample>_spatialTable.csv       # per-barcode spatial counts
    ├── <sample>_bcSpatial*.png         # barcode-density spatial heatmaps
    ├── <sample>_plate_map.html/.png, _pareto.html, _barplot.html, _denseOnOff.html/.png
    ├── <sample>_hiLoWarn.csv/.png      # high / low outlier lane warnings
    └── clean_v1_outputs/, clean_v2_*_outputs/   # detection / cleaning variants

File	Description
<sample>_report.html	Interactive barcode QC report — match rates, spatial distribution, and warnings.
<sample>_cleanTable.csv	The correction table (mer, finalDownFract) used as the preprocessing ds_table for cleaning.
<sample>_spatialTable.csv	Per-barcode spatial counts.
<sample>_bcSpatial*.png	Barcode-density spatial heatmaps (raw, 2× mean, 95th percentile).
<sample>_hiLoWarn.csv / .png	Flagged high / low outlier rows and columns.
clean_v1_outputs/, clean_v2_*_outputs/	Per-variant detection / cleaning intermediates.

CLI

The same barcode QC can be run locally via the atlasxomics/barcodeqc CLI — see the DIY / Local Processing page.

Example run

(Representative LaunchPlan / batch-table example to be added.)