None of the metadata files attached to this study include the custom single cell annotations referenced in the study (https://www.jci.org/articles/view/185217). Can they please be added?

Created by Noah S User217
Hi @joshuabrand, Thank you for your interest in the [AMP SLE PBMC CITE-seq](https://arkportal.synapse.org/Explore/Datasets/DetailsPage?id=75065673) data. This dataset is an "experimental" dataset and is thus in a more raw state (https://help.arkportal.org/help/navigating-the-portal#Explore). The deeper cell type predictions you are looking for are typically included in more processed forms of data that are provided in "publication" datasets. The authors of the linked pre-print are nearing publication of a peer-reviewed version of their manuscript - once their study is accepted for publication an corresponding publication dataset will be available in ARK. Sincerely, Jess Vera Senior Biomedical Data Manager Sage Bionetworks
Hi @jmvera, I noticed that this was also the case for the phase II SLE PBMC CITE-Seq data linked to this preprint: https://www.biorxiv.org/content/10.1101/2025.08.11.669754v3. The synapse id for this data is: syn74758633. I was wondering if there is a planned future release that includes deeper cell annotations? I've downloaded the rds files: gex, adt, and metadata, The best I can see are broad cell typing: table(ampdata_meta$broadCellType) B/Plasma Cell DC Monocyte NK/ILC T Cell 81948 11791 191030 78749 209839 Thanks!
I was able to access their annotations via the new RDS file. Thank you for the quick resolution.
Hi @User217, The data contributor was able to provide the predicted cell type annotations. Version 3 of [`CITEseq_cell_meta.rds`](https://www.synapse.org/Synapse:syn64377383.3) now includes the column `pred_cluster`. A total of 70 unique cell types labels are listed. Please note - of 488,540 cells described in this data.frame, only 289,361 cells have a predicted cell type as these are the subset of cells that were confidently assigned to one of the major immune lineages (T cells, B cells, myeloid cells, or NK cells) based on the expression of key, broad cell-type markers. Cells with `pred_cluster = NA` are ambiguous populations that could not be reliably classified into one of these major lineages and were therefore excluded from downstream analyses described in the corresponding publication.
Thank you for the quick response and for reaching out to the data contributors. I glimpsed inside the metadata of all 16 RDS files and did not find a cell annotation column.
Hi @User217, Thank you for your interest in this data. We can reach out to the data contributors for clarification. In the meantime, please note that there are several SeuratObjects saved to Rds files included in this dataset. It is standard to store cell annotations to the column metadata data.frame saved in the SeuratObject. We recommend reviewing the SeuratObjects to see if this information is available there.

Custom Cell Annotations Not Present in Metadata Files page is loading…