Processed data for X-Atlas/Orion: Genome-wide Perturb-seq Datasets via a Scalable Fix-Cryopreserve Platform for Training Dose-Dependent Biological Foundation Models

<p dir="ltr">This dataset (X-Atlas/Orion) contains processed data from two genome-wide Perturb-seq experiments in HCT116 and HEK293T cell lines described in the <a href="https://www.biorxiv.org/content/10.1101/2025.06.11.659105v1" rel="noreferrer" target="_blank">manuscript</a> <b>X-Atlas/Orion: Genome-wide Perturb-seq Datasets via a Scalable Fix-Cryopreserve Platform for Training Dose-Dependent Biological Foundation Models</b></p><p dir="ltr">Dataset:</p><ul><li>HCT116:</li><li><ol><li><b>HCT116_filtered_dual_guide_cells.h5ad</b>: HCT116 cells that contain two sgRNAs targeting the same gene and from the same guide pair</li><li><b>HCT116_filtered_dual_guide_cells.h5ad.md5</b>: checksum for HCT116_filtered_dual_guide_cells.h5ad</li></ol></li><li>HEK293T:</li><li><ol><li><b>HEK293T_filtered_dual_guide_cell</b><b>s</b><b>.h5ad</b>: HEK293T cells that contain two sgRNAs targeting the same gene and from the same guide pair</li><li><b>HEK293T_filtered_dual_guide_cell</b><b>s</b><b>.h5ad.md5</b>: checksum for HEK293T_filtered_dual_guide_cells.h5ad</li></ol></li></ul><p dir="ltr">h5ads containing all aligned cells to be released at a later date.</p><p dir="ltr">Description of h5ads: h5ads are <a href="https://anndata.readthedocs.io/en/latest/" rel="noreferrer" target="_blank">AnnData</a> objects that contain the following metadata</p><ul><li>cell-level (obs):</li><li><ul><li>sample: GEM batch</li><li>num_features: number of guides</li><li>guide_target: guide identity</li><li>gene_target: gene targeted by guide</li><li>n_genes_by_counts: number of genes with non-zero counts</li><li>total_counts: total UMIs</li><li>total_counts_mt: total UMIs from MT genes</li><li>pct_counts_mt: % UMIs from MT genes</li><li>pass_guide_filter: boolean if cells contains two guides from the same guide pair</li></ul></li><li>gene-level (var):</li><li><ul><li>mt: boolean if gene is MT gene</li><li>n_cells_by_counts: number of cells gene has non-zero UMIs in</li><li>mean_counts: mean UMIs over all cells</li><li>pct_dropout_by_counts: % of cells this gene does not appear in</li><li>total_counts: sum of UMIs for a gene</li></ul></li></ul><p dir="ltr">Other files:</p><ul><li> <b>guide_library.csv</b>: Table containing guide pairs in X-Atlas/Orion. Guide sequences are from <a href="https://elifesciences.org/articles/81856" rel="noreferrer" target="_blank">Replogle, et al. eLife (2022)</a>. Description of columns:</li><li><ul><li>target_gene: gene symbol of target gene</li><li>target_gene_id: Ensembl ID of target gene</li><li>id_a: unique identifier for the first guide in the pair (Guide A). Used in .obs.guide_target</li><li>id_a (Replogle et al): original name of Guide A in Replogle, et al. eLife (2022)</li><li>sequence_a: sequence of Guide A </li><li>id_b: unique identifier for the second guide in the pair (Guide B). Used in .obs.guide_target</li><li>id_b (Replogle et al): original name of Guide B in Replogle, et al. eLife (2022)</li><li>sequence_b: sequence of Guide B</li><li>id_ab: unique identifier for guide pair (id_a | id_b) </li></ul></li></ul><p></p>