Update test-data to work with latest version of pixelator (0.29.0)

new-test-data/README.md

@@ -0,0 +1,7 @@
+# Pixelator 0.29 test data overlay
+
+Updated fixtures for pixelator 0.29.0 live under `new-test-data/pna/`.
+Unchanged files are copied from `testdata/pna/`; changed or new files are stored here.
+
+The v2 samplesheet with the prerelease panel is at
+`new-test-data/samplesheet/pna/samplesheet_proxiome_v2.csv`.

new-test-data/pna/README.md

@@ -0,0 +1,50 @@
+## Test datasets origin
+
+This test data is generated by Pixelgen Technologies AB for the nf-core/pixelator pipeline.
+
+Synthetic PNA test data in this folder was generated with internal prototype scripts.
+
+### FASTQ generation procedure
+
+Data generation used deterministic seeds for reproducibility.
+
+Generation was performed in two steps:
+
+1. Build a synthetic edgelist parquet with:
+
+- 20 cells
+- 600 nodes per cell
+- 22200 edges in total (200 crossing edges)
+- 20% of antibodies are hashing antibodies
+
+2. Convert each parquets into paired FASTQ files and using the corresponding PNA panel definitions:
+
+- proxiome-v1-immuno-155-v1.1 panel for `sample1`, `sample1_reseq`, `sample2`
+- proxiome-v2-immuno-155-prerelease panel for `pool1`, `pool1_reseq`, `pool2`
+
+This generates the files in `testdata/pna/fastq`, including:
+
+- `sample1*_R1.fastq.gz`, `sample1*_R2.fastq.gz`, `sample2_R1.fastq.gz`, `sample2_R2.fastq.gz`
+- `pool1*_R1.fastq.gz`, `pool1*_R2.fastq.gz`, `pool2_R1.fastq.gz`, `pool2_R2.fastq.gz`
+- corresponding `*.metadata.json` files
+
+### Module output generation (`testdata/pna/modules`)
+
+Module artifacts were then generated with Pixelator `single-cell-pna` subcommands, in this order:
+
+1. `amplicon`
+2. `demux`
+3. `collapse` + `combine-collapse`
+4. `graph`
+5. `denoise` (pool-level for proxiome-v2)
+6. `sample-calling`
+7. `layout`
+8. `analysis`
+
+Representative options (from module metadata):
+
+- `demux`: `--design proxiome-v2 --panel proxiome-v2-immuno-155-prerelease`
+- `sample-calling`: `--samplesheet samplesheet_proxiome_v2.csv --enrichment-threshold 10.0`
+- `graph`: `--component-size-min-threshold 100 --multiplet-recovery`
+- `layout`: `--layout-algorithm coarsened_pmds_3d`
+- `analysis`: `--compute-proximity --proximity-nbr-of-permutations 100`