Feature barcode assignment: 10x TotalSeq-B public lane

• Category: Public data

• Assay type: feature_barcode

• Evidence level: Documented example

• Dataset: 10x Genomics TotalSeq-B antibody Feature Barcode R2 sample

• Condition shown: Per-read fixed-window feature-barcode assignment with exact-slice baseline agreement.

• Primary report: docs/benchmarks/feature_barcode/README.md

What This Shows

• DotMatch can perform fixed-window per-read feature-barcode assignment on a public 10x-style lane.

• The report keeps per-read assignment semantics separate from cell/UMI quantification.

What It Does Not Show

• This is not Cell Ranger feature-count replacement evidence.

• Cell barcode, UMI aggregation, and cell-level feature quantification are outside this lane.

Comparator And Validation

Comparator semantics: The public lane uses the 10x feature reference fixed R2 pattern and a transparent exact-slice hash baseline for per-read fixed-window assignment.

Validation: The public gate requires synthetic diagnostics, public DotMatch k=0 and k=1 rows, zero validation mismatches, and exact agreement between DotMatch k=0 and the exact-slice baseline.

Report Examples

• Feature-barcode benchmark report (markdown)

Raw Artifacts

• benchmarks/raw/feature_barcode.csv

• examples/feature_barcode/data/metadata.json

Reproduce Or Check

• make bench-feature-barcode-public

• make feature-barcode-public-gate