File size: 3,030 Bytes
3040359 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 | # CausalBench K562 and RPE1
Filtered single-cell CRISPRi data from [Replogle et al. (2022)](https://doi.org/10.25452/figshare.plus.20029387.v1), prepared following [CausalBench](https://doi.org/10.1038/s42003-025-07764-y). Data license: CC BY 4.0, credited to Replogle et al.
| Dataset | Cells | Genes | Prepared table |
| --- | ---: | ---: | --- |
| K562 | 162,751 | 622 | [causalbench_k562.tsv.gz](data/causalbench_k562.tsv.gz) |
| RPE1 | 162,733 | 383 | [causalbench_rpe1.tsv.gz](data/causalbench_rpe1.tsv.gz) |
Each gzip TSV contains gene-expression columns (Ensembl IDs) and an `intervention` column: a targeted gene ID, `non-targeting` for controls, or `excluded` for targets not retained by preprocessing. `excluded` cells are not controls.
## Sources and preprocessing
The downloaded inputs were:
- `k562.h5ad`: [K562 count matrix from Figshare, file 35773219](https://ndownloader.figshare.com/files/35773219).
- `rpe1.h5ad`: [RPE1 count matrix from Figshare, file 35775606](https://ndownloader.figshare.com/files/35775606).
- `summary_stats.xlsx`: [Replogle et al. supplementary workbook (`1-s2.0-S0092867422005979-mmc2.xlsx`)](https://ars.els-cdn.com/content/image/1-s2.0-S0092867422005979-mmc2.xlsx), using sheets `TabB_K562_day6_summary_stat` and `TabC_RPE1_summary_statistic`.
The names above are the local filenames used during preparation.
The reference is the authors' [preprocessing implementation at commit 1a2143c](https://github.com/causalbench/causalbench/blob/1a2143cffdc85f835b41ce8d52034be1bf903e71/causalscbench/data_access/utils/preprocessing.py), with strong-perturbation filtering enabled:
1. Keep controls and perturbations with >50 differentially expressed genes, knockdown <= -0.3, and >25 filtered cells in the summary.
2. For measured targets, keep perturbed cells whose target expression is at or below the control 10th percentile.
3. Remove zero-total cells, normalize cell totals to the retained-cell median using all original genes, and apply natural `log1p`.
4. Keep measured genes targeted in >100 retained cells; relabel unsupported targets `excluded` without dropping their rows.
K562 used the original function. RPE1 used a memory-saving implementation of the same rules. [Raw inputs, preparation code, and environment](https://huggingface.co/datasets/huggingbahl21/causalbench) are available for reproduction.
## Comparison with CausalBench
These tables use the original input datasets and follow the authors' filtered preprocessing linked above. They are our prepared exports, not processed files downloaded from the CausalBench authors. An exact match to author-provided processed matrices has not been verified.
Both downloaded count matrices matched Figshare's checksums. Rebuilding from the raw inputs reproduced the saved expression arrays and gene/intervention labels; exporting to TSV changed numerical values by at most `2.3841e-7`. The [preparation code and inputs](https://huggingface.co/datasets/huggingbahl21/causalbench) are available to reproduce the tables.
|