This repo provides the data, code, and website for our work on "Cellular deconstruction of inflamed synovium defines diverse inflammatory phenotypes in rheumatoid arthritis". The data is generated through a collaborative effort with NIH funded AMP RA/SLE (Accelerating Medicines Partnership Rheumatoid Arthritis and Systemic Lupus Erythematosus).
- RA Synovial Single-cell Multimodal (transcriptomics and Proteomics) Cell Atlas
- Inflammatory tissue stratification: RA Synovial Cell Type Abundance Phenotypes (CTAPs)
- Associations of certain CTAPs with disease-relevant cytokines, histology, serology metrics
- Reveal significant associations between RA causal genes and cell type-specific synovial CTAPs
- Infer inflammatory phenotypes from clinical trial bulk RNA-seq of RA synovial tissue
- Functional cell-cell interaction and immune mediator assays
- Microscopy analysis for different synovial phenotypes
Our paper can be cited: Zhang*, Jonsson*, Nathan*, Wei*, Millard*, et al, Nature, 2023, https://www.nature.com/articles/s41586-023-06708-y
Feel free to check out our Website regarding the single-cell data and results.
Source code for reproducing the results are available at our Github repo.
CITE-seq single-cell expression matrices and sequencing and bulk expression matrices are available on Synapse (https://doi.org/10.7303/syn52297840).
Associated genotype and clinical data are available through the Arthritis and Autoimmune and Related Diseases Knowledge Portal (ARK Portal, https://arkportal.synapse.org/Explore/Datasets/DetailsPage?id=syn52297840).
- Each cell type-specific reference is provided based on one cell type-focused analysis. This could support reference mapping analysis based on the anchor genes, a downstream analysis to query and annotate cells from other disease contexts.
Bcell_reference.rds
Bcell_uwot_model
NK_reference.rds
NK_uwot_model
T_reference.rds
T_uwot_model
endothelial_reference.rds
endothelial_uwot_model
fibroblast_reference.rds
fibroblast_uwot_model
myeloid_reference.rds
myeloid_uwot_model
Every reference data has the same data structure format. Taking one cell type reference as an example, you will see the substructures as follows.
> ref <- readRDS("myeloid_reference_2023-03-12.rds")
> str(ref)
$ meta_data :'data.frame': 76181 obs. of 7 variables:
..$ cell : chr [1:76181] "BRI-399_AAACCCAGTAGGAGGG" "BRI-399_AAACGCTGTTCAAGTC" "BRI-399_AAAGGATTCTGTACAG" "BRI-399_AAAGGTAAGCTGGCTC" ...
..$ sample : chr [1:76181] "BRI-399" "BRI-399" "BRI-399" "BRI-399" ...
..$ cluster_number: chr [1:76181] "M-10" "M-10" "M-0" "M-7" ...
..$ cluster_name : chr [1:76181] "M-10: DC2" "M-10: DC2" "M-0: MERTK+ SELENOP+ LYVE1+" "M-7: IL1B+ FCN1+ HBEGF+" ...
..$ nUMI : num [1:76181] 33644 17084 4999 16721 11235 ...
..$ nGene : int [1:76181] 5515 3806 1770 3767 3101 4788 3460 4028 3463 3152 ...
..$ percent_mito : num [1:76181] 0.0847 0.0446 0.1658 0.1576 0.0129 ...
$ vargenes : tibble [3,547 × 3] (S3: tbl_df/tbl/data.frame)
..$ symbol: chr [1:3547] "CXCL13" "CCL17" "CHI3L1" "SPP1" ...
..$ mean : Named num [1:3547] 0.0256 0.0208 0.0429 1.0576 0.2033 ...
.. ..- attr(*, "names")= chr [1:3547] "CXCL13" "CCL17" "CHI3L1" "SPP1" ...
..$ stddev: Named num [1:3547] 0.227 0.226 0.192 1.682 0.452 ...
.. ..- attr(*, "names")= chr [1:3547] "CXCL13" "CCL17" "CHI3L1" "SPP1" ...
$ loadings : num [1:3547, 1:20] -0.00166 -0.00647 0.00709 0.00111 0.04714 ...
$ R : num [1:100, 1:76181] 2.21e-14 1.81e-06 4.35e-08 3.66e-08 1.70e-14 ...
$ Z_orig : num [1:20, 1:76181] -7.567 0.538 -8.644 -5.862 -2.298 ...
$ Z_corr : num [1:20, 1:76181] -7.559 -1.46 -5.743 -3.853 0.223 ...
..- attr(*, "dimnames")=List of 2
.. ..$ : chr [1:20] "harmony_1" "harmony_2" "harmony_3" "harmony_4" ...
.. ..$ : chr [1:76181] "4" "28" "38" "47" ...
$ centroids : num [1:20, 1:100] 0.92797 0.29046 -0.05318 -0.00418 -0.00714 ...
$ cache :List of 2
..$ : num [1:100, 1] 768 865 845 756 775 ...
..$ : num [1:100, 1:20] 7880 -11077 -6132 -1873 6180 ...
$ umap :List of 1
..$ embedding: num [1:76181, 1:2] -5.02 -5.62 1.78 -4.26 -4.35 ...
.. ..- attr(*, "scaled:center")= num [1:2] 0.19 0.343
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : NULL
.. .. ..$ : chr [1:2] "UMAP1" "UMAP2"
$ save_uwot_path: chr "./myeloid_uwot_model_2021-04-29"
As you can see above, the cluster_number and cluster_name are given under ref$meta_data. Further, some key outpus are explained as follows:
# vargenes: variable genes, means, and standard deviations used for scaling
# loadings: gene loadings for projection into pre-Harmony PC space
# R: Soft cluster assignments
# Z_orig: Pre-Harmony PC embedding
# Z_corr: Harmonized PC embedding
# centroids: locations of final Harmony soft cluster centroids
# cache: pre-calculated reference-dependent portions of the mixture model
# umap: UMAP coordinates
# save_uwot_path: path to saved uwot model (for query UMAP projection into reference UMAP coordinates)
- CTAP assignment to donor ID:
CTAP_donor_mapping.xlsx
> map <- read_excel("CTAP_donor_mapping.xlsx")
> head(map)
subject_id donor CTAP
1 300-0310 BRI-405 E + F + M
2 300-0309 BRI-411 E + F + M
3 300-0174 BRI-479 E + F + M
4 300-0175 BRI-525 E + F + M
5 300-0529 BRI-554 E + F + M
6 300-0145 BRI-589 E + F + M
- Reference for all cell type integrative analysis:
all_cells_reference.rds
all_cells_uwot_model
- Cluster annotations for fine-grained cell states:
fine_cluster_all_314011cells_82samples.rds
'data.frame': 314011 obs. of 5 variables:
$ sample : chr "BRI-399" "BRI-399" "BRI-399" "BRI-399" ...
$ cell : chr "BRI-399_AAACGAACAGTCTGGC" "BRI-399_AAAGGATGTCTCAAGT" "BRI-399_AAAGTGACATCGAACT" "BRI-399_AAAGTGAGTGCACAAG" ...
$ cluster_number: chr "B-2" "B-1" "B-2" "B-1" ...
$ cluster_name : Factor w/ 77 levels "B-0: CD24+CD27+CD11b+ switched memory",..: 2 4 2 4 1 4 3 3 2 8 ...
$ cell_type : chr "B cell" "B cell" "B cell" "B cell" ...
- Raw and processed matrices:
raw_mRNA_count_matrix.rds: Raw mRNA count
raw_protein_count_matrix.rds: Raw protein count
qc_mRNA_314011cells_log_normalized_matrix.rds: QCed mRNA normalized data matrix
qc_protein_CLR_normalized_filtered_matrix.rds: QCed protein normalized data matrix
Please email Fan Zhang ([email protected]) and Helena Jonsson ([email protected]) for any questions.