Error when using Fibroblast atlas

[jk86754]

Hi,

I am trying to use the pre-built fibroblast atlas (https://sandbox.zenodo.org/record/772596#.YdRr5mDMIuV) in Symphony. I read in the RDS file as specified:

fibroblast_reference <- readRDS('fibroblast_atlas.rds')

When trying to plot the reference

p = plotReference(fibroblast_reference,
as.density = TRUE, # plot density or individual cells
bins = 14, # if density, nbins parameter for stat_density_2d
bandwidth = 1, # if density, bandwidth parameter for stat_density_2d
title = "Symphony Reference", # Plot title
color.by = 'Cluster', # metadata column name for cell type labels
show.legend = TRUE, # Show cell type legend
show.labels = TRUE, # Show cell type labels
show.centroids = FALSE) # Plot soft cluster centroid locations)

I get an error:

Error in plotReference(ref_pbmcs, as.density = TRUE, bins = 14, bandwidth = 1, :
Error: umap slot is empty. UMAP was not saved for this reference!

Any help would be appreciated!

sessionInfo()
R version 4.0.4 (2021-02-15)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)

Matrix products: default
BLAS/LAPACK: /usr/lib64/libopenblasp-r0.3.3.so

locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 LC_PAPER=en_US.UTF-8
[8] LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] patchwork_1.1.1 RColorBrewer_1.1-2 ggrastr_1.0.1 ggthemes_4.2.4 plyr_1.8.6 Matrix_1.4-0 matrixStats_0.61.0 forcats_0.5.1
[9] stringr_1.4.0 dplyr_1.0.7 purrr_0.3.4 readr_2.1.1 tidyr_1.1.4 tibble_3.1.6 ggplot2_3.3.5 tidyverse_1.3.1
[17] singlecellmethods_0.1.0 data.table_1.14.2 symphony_0.1.0 harmony_0.1.0 Rcpp_1.0.7

loaded via a namespace (and not attached):
[1] fs_1.5.2 usethis_2.1.5 lubridate_1.8.0 devtools_2.4.3 httr_1.4.2 rprojroot_2.0.2 tools_4.0.4 backports_1.4.1 utf8_1.2.2 R6_2.5.1
[11] irlba_2.3.5 vipor_0.4.5 uwot_0.1.11 DBI_1.1.2 colorspace_2.0-2 withr_2.4.3 tidyselect_1.1.1 prettyunits_1.1.1 processx_3.5.2 curl_4.3.2
[21] compiler_4.0.4 cli_3.1.0 rvest_1.0.2 xml2_1.3.3 desc_1.4.0 scales_1.1.1 callr_3.7.0 pkgconfig_2.0.3 sessioninfo_1.2.2 dbplyr_2.1.1
[31] fastmap_1.1.0 rlang_0.4.12 readxl_1.3.1 rstudioapi_0.13 generics_0.1.1 jsonlite_1.7.2 magrittr_2.0.1 ggbeeswarm_0.6.0 munsell_0.5.0 fansi_0.5.0
[41] lifecycle_1.0.1 stringi_1.7.6 pkgbuild_1.3.1 grid_4.0.4 ggrepel_0.9.1 crayon_1.4.2 lattice_0.20-45 haven_2.4.3 cowplot_1.1.1 hms_1.1.1
[51] ps_1.6.0 pillar_1.6.4 codetools_0.2-18 pkgload_1.2.4 reprex_2.0.1 glue_1.6.0 remotes_2.4.2 modelr_0.1.8 vctrs_0.3.8 tzdb_0.2.0
[61] testthat_3.1.1 cellranger_1.1.0 gtable_0.3.0 RANN_2.6.1 assertthat_0.2.1 cachem_1.0.6 broom_0.7.11 beeswarm_0.4.0 memoise_2.0.1 ellipsis_0.3.2

[jk86754]

If I understand your description correctly, I also need the uwot model file. This seems to be missing from what is available on Zenodo.

[jk86754]

Sorry, one more note: I also cannot find meta data associated with the Fibroblast atlas.

[joycekang]

Hi @jk86754 ,
Thanks for your message! Currently, the fibroblast atlas supports mapping to the atlas and label transfer (see https://github.com/immunogenomics/fibroblastatlas for tutorial). The fibroblast atlas may have been generated with a slightly different internal structure than the other Symphony atlases in terms of the way the uwot model is saved.

I have reached out to @ilyakorsunsky who generated the fibroblast atlas, and he should be able to help. Note that some of the information (e.g. cell metadata) might only be made public once the paper is officially published (in a journal).