screfmapping

Overview

"screfmapping" is a pipeline that facilitates the extraction of CD4+ T cells from the single-cell RNA-seq raw data of peripheral blood mononuclear cells (PBMCs) and maps them to our annotated clusterL1, L2 atlas. The Azimuth pipeline is employed to extract CD4+ T cells, and Symphony mapping, which includes batch effect adjustment by Harmony, is used for mapping to our atlas.

We've included an example analysis in example.R.
Below, we provide an overview of the function.

Function

# 1st step, CD4T extraction
extract_cells_seuratobj(query = q,                   # query_SeuratObject
                        reference = reference,       # Azimuth_reference
                        prefix = prefix)             # output_file_path

# 2nd step, label transfer
reference_mapping_seuratobj(ref = ref,               # our_annotated_clusterL1,L2_data
                            query_obj = query_obj,   # extracted_CD4T_SeuratObject_with or without_metadata
                            prefix = prefix)          # output_file_path

Rscript

docker run --rm -it -v ${PWD}:/home/rstudio/autoimmune_10x  yyasumizu/screfmapping:0.0.1 Rscript example.R

Pre-annotated reference data

Required files for CD4T classifications are included in the Docker image (Docker hub: yyasumizu/screfmapping:0.0.1). Users can also download the ref_Reference_Mapping_20220525.RData file from here. In that case, place it in the /screfmapping/data/ directory.　
Users will need this file for ref in the reference_mapping_seuratobj function.

Python route (symphonypy)

The pipeline above is the R implementation (Seurat + SCTransform + Symphony) for the CD4+ T cell atlas. A Python implementation is in python/, covering both the CD4+ T cell and B cell atlases, with an end-to-end notebook for each.

If you map with symphonypy, read python/README.md first. The AnnData references were built with regress_out([total_counts, pct_counts_mt, S_score, G2M_score]) before scaling, so the query has to be put through the same covariate regression using the coefficients the reference ships (varm['regress_beta']). python/screfmapping_symphonypy.py does this for you. The R route on this page is unaffected: SCTransform scale.data is computed the same way for reference and query.

Output

extract_cells_seuratobj

• ${prefix}_CD4T_MetaData.rds
• ${prefix}_CD4T_AssayData.rds : Input_data_for_ReferenceMapping

reference_mapping_seuratobj

• ${prefix}_predict_clusterL1L2_Reference_Mapping.pdf
• ${prefix}_Reference_Mapping.csv : Symphony result

Tips

The number of neighbors (k) to use when finding anchors

Our "screfmapping" is expected to be used for PBMC datasets. However, some people may want to use it for CD4+ T cell-enriched datasets. In such cases, we have noticed that a proportion of CD4+ T cells tend to be misannotated as non-CD4+ T cells (approximately 4%). Empirically, we found that terminally differentiated effector memory T cells (Temra) tended to be annotated as CD8+ T cells or NK cells because those transcriptomes were similar.
Users may be able to deal with this issue by optimizing the k.anchor values of the FindTransferAnchors in the extract_cells_seuratobj function in ref_mapping_seuratobj.R. The lower k.anchor values (for example, k.anchor = 3, compared to the default k.anchor = 5) worked well for CD4+ T cell-enriched datasets.

# lines 40-52 (in `ref_mapping_seuratobj.R`) should be replaced as below.

anchors <- FindTransferAnchors(reference = reference$map,
                               query = query,
                               k.anchor = 3, # change here
                               k.filter = NA,
                               reference.neighbors = "refdr.annoy.neighbors",
                               reference.assay = "refAssay",
                               query.assay = "refAssay",
                               reference.reduction = "refDR",
                               normalization.method = "SCT",
                               features = intersect(rownames(x = reference$map),
                                                    VariableFeatures(object = query)),
                               dims = 1:50,
                               n.trees = 20,
                               mapping.score.k = 100)

Citation

Yasumizu, Y., Takeuchi, D., Morimoto, R., Takeshima, Y., Okuno, T., Kinoshita, M., Morita, T., Kato, Y., Wang, M., Motooka, D., et al. (2024). Single-cell transcriptome landscape of circulating CD4+ T cell populations in autoimmune diseases. Cell Genomics.
https://doi.org/10.1016/j.xgen.2023.100473

Licence

This software is freely available for academic users. Usage for commercial purposes is not allowed. Please refer to the LICENCE page.