Last updated: 2025-06-23

Checks: 6 1

Knit directory: single-cell-jamboree/analysis/

This reproducible R Markdown analysis was created with workflowr (version 1.7.1). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.

R Markdown file: uncommitted changes

The R Markdown file has unstaged changes. To know which version of the R Markdown file created these results, you’ll want to first commit it to the Git repo. If you’re still working on the analysis, you can ignore this warning. When you’re finished, you can run wflow_publish to commit the R Markdown file and build the HTML.

Environment: empty

Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.

Seed: set.seed(1)

The command set.seed(1) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.

Session information: recorded

Great job! Recording the operating system, R version, and package versions is critical for reproducibility.

Cache: none

Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.

File paths: relative

Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.

Repository version: 2e7cdf0

Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.

The results in this page were generated with repository version 2e7cdf0. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.

Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated: 


Untracked files:
    Untracked:  data/GSE156175_RAW/
    Untracked:  data/panc_cyto_lsa_tm_k12.rds
    Untracked:  data/pancreas_cytokine_lsa.Rdata
    Untracked:  output/panc_cyto_lsa_res/

Unstaged changes:
    Modified:   analysis/pancreas_cytokine_mf_comparison.Rmd

Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.

These are the previous versions of the repository in which changes were made to the R Markdown (analysis/pancreas_cytokine_mf_comparison.Rmd) and HTML (docs/pancreas_cytokine_mf_comparison.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.

File Version Author Date Message
html 2e7cdf0 Peter Carbonetto 2025-06-23 Updated the Structure plots in the pancreas_cytokine_mf_comparison analysis.
Rmd e7ffd05 Peter Carbonetto 2025-06-23 wflow_publish("pancreas_cytokine_mf_comparison.Rmd", verbose = T,
Rmd 546d291 Peter Carbonetto 2025-06-23 Improved the structure plots for the topic model with k=12 in the pancreas_cytokine_mf_comparison analysis.
Rmd 7af0ecd Eric Weine 2025-06-17 added umap plot to clustering
Rmd ba07b39 Eric Weine 2025-06-16 added pancreas cytokine mf comparison
html ba07b39 Eric Weine 2025-06-16 added pancreas cytokine mf comparison

This is our initial NMF analysis of ther Stancill 2021 LSA data set.

Load the packages used in this analysis:

library(Matrix)
library(readr)
library(dplyr)
library(fastTopics)
library(log1pNMF)
library(ggplot2)
# Warning: package 'ggplot2' was built under R version 4.4.1
library(cowplot)

Load the data set:

load("../data/pancreas_cytokine_lsa.Rdata")
barcodes   <- as.data.frame(barcodes)
clusters   <- factor(barcodes$celltype,
                     c("Acinar","Ductal","Endothelial/Mesnchymal","Macrophage",
                       "Alpha","Beta","Delta","Gamma"))
conditions <- factor(barcodes$condition,
                     c("Untreated","IL-1B","IFNg","IL-1B_IFNg"))

Fit the log1p models:

cc_vec <- c(1)

K_vec <- c(12, 15)

init_method_vec <- c("rank1")

for (init_method in init_method_vec) {

  for (K in K_vec) {

    for (cc in cc_vec) {

      set.seed(1)
      fit <- fit_poisson_log1p_nmf(
        Y = counts,
        K = K,
        cc = cc,
        init_method = init_method,
        loglik = "exact",
        control = list(maxiter = 250)
      )

      readr::write_rds(
        fit,
        glue::glue("results/lsa_pancreas_cytokine_log1p_c{cc}_{init_method}_init_K{K}.rds")
      )

    }

  }

}

Fit the topic models:

set.seed(1)
tm_k12 <- fit_poisson_nmf(
  X = as(counts, "CsparseMatrix"),
  k = 12,
  control = list(nc = 7)
)

set.seed(1)
tm_k15 <- fit_poisson_nmf(
  X = as(counts, "CsparseMatrix"),
  k = 15,
  control = list(nc = 7)
)

Topic model, K = 12

Here, I attempt to separate out fitted topics based on their association with either celltype or treatment.

tm_k12 <- read_rds("../output/panc_cyto_lsa_res/tm_k12.rds")

The topic model appears to have more factors that are specific to celltype, and fewer that seem to capture the treatment effects.

set.seed(1)
celltype_topics <- c(2:3,5:11)
other_topics <- c(1,4,12)
topic_colors <- fastTopics:::kelly()[c(2:12,14)]
i <- c(sample(which(clusters == "Beta"),800),
       which(clusters != "Beta"))
L <- poisson2multinom(tm_k12)$L
p1 <- structure_plot(L[i,],grouping = clusters[i],gap = 15,n = Inf,
                     topics = celltype_topics,colors = topic_colors)$plot   
p2 <- structure_plot(L[i,],grouping = clusters[i],gap = 15,n = Inf,
                     topics = other_topics,colors = topic_colors)$plot
p3 <- structure_plot(L[i,],grouping = conditions[i],gap = 20,n = Inf,
                     topics = other_topics,colors = topic_colors)$plot
plot_grid(p1,p2,p3,nrow = 3,ncol = 1,rel_heights = c(3.3,3,2))

Version Author Date
2e7cdf0 Peter Carbonetto 2025-06-23

Topic 1 appears to be non-IL-1B specific, whereas topic 4 appears to be a general cytokine effect (but perhaps moreso capturing the effects of IL-1B).

Poisson log1p NMF, K = 12

Let’s now look at the results from the Poisson log1p NMF, with \(K = 12\).

log1p_k12 <- read_rds(file.path("../output/panc_cyto_lsa_res",
               "lsa_pancreas_cytokine_log1p_c1_rank1_init_K12.rds"))

Here are the factors that appeared most correlated with cell type.

scale_cols <- function (A, b)
  t(t(A) * b)
set.seed(1)
celltype_topics = c(1,2,3,5,6,7,8,9)
topic_colors <- fastTopics:::kelly()[c(2:12,14)]
L <- log1p_k12$LL
d <- apply(L,2,max)
L <- scale_cols(L,1/d)
p1 <- structure_plot(L[i,],grouping = clusters[i],gap = 15,n = Inf,
                     topics = celltype_topics,colors = topic_colors)$plot +
  labs(y = "membership",fill = "")                   
p2 <- structure_plot(L[i,],grouping = clusters[i],gap = 15,n = Inf,
                     topics = c(4,10,11,12),colors = topic_colors)$plot +
  labs(y = "membership",fill = "")
p3 <- structure_plot(L[i,],grouping = conditions[i],gap = 30,n = Inf,
                     topics = c(4,10,11,12),colors = topic_colors)$plot +
  labs(y = "membership",fill = "")                   
plot_grid(p1,p2,p3,nrow = 3,ncol = 1)

Version Author Date
2e7cdf0 Peter Carbonetto 2025-06-23

NOTE: UPDATE THIS TEXT. The pockets of purple and yellow in the beta cell cluster may represent cells that were misclassified (though it is very hard to tell). Here, it appears that the pink factor is specific to conditions that are not treated with IL-1B, where the blue factor is specific to conditions treated with IL-1B. The orange factor appears specific to conditions treated with IFNg. The condition with both treatments has both the orange and the blue factors.

K = 15

tm_fit15 <- readr::read_rds("../data/panc_cyto_lsa_res/tm_k15.rds")
structure_plot(tm_fit15, grouping = barcodes$celltype, gap = 20)

# Treatment specific factors:
# k2
# k8
# k9
# k10
# k12
# k14
# k15

# celltype specific factors:
# k1
# k3
# k4
# k5
# k6
# k7
# k11
# k13
structure_plot(
  tm_fit15, grouping = barcodes$condition, gap = 20,
  topics = paste0("k", c(2, 8, 9, 10, 12, 14, 15))
  )

Here, it appears that the pink factor (k10) and grey (k8) is mostly specific to the untreated condition, the dark orange (k14) is specific to IL-1B, the maroon (k15) is represented in all treated conditions.

log1p_fit15 <- readr::read_rds("../data/panc_cyto_lsa_res/lsa_pancreas_cytokine_log1p_c1_rank1_init_K15.rds")
normalized_structure_plot(log1p_fit15, grouping = barcodes$celltype, gap = 20, n = Inf)

sessionInfo()
# R version 4.4.0 (2024-04-24)
# Platform: aarch64-apple-darwin20
# Running under: macOS Ventura 13.5
# 
# Matrix products: default
# BLAS:   /Library/Frameworks/R.framework/Versions/4.4-arm64/Resources/lib/libRblas.0.dylib 
# LAPACK: /Library/Frameworks/R.framework/Versions/4.4-arm64/Resources/lib/libRlapack.dylib;  LAPACK version 3.12.0
# 
# locale:
# [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
# 
# time zone: America/New_York
# tzcode source: internal
# 
# attached base packages:
# [1] stats     graphics  grDevices utils     datasets  methods   base     
# 
# other attached packages:
# [1] cowplot_1.1.3     ggplot2_3.5.2     log1pNMF_0.1-4    fastTopics_0.7-24
# [5] dplyr_1.1.4       readr_2.1.5       Matrix_1.7-0     
# 
# loaded via a namespace (and not attached):
#  [1] gtable_0.3.6        xfun_0.52           bslib_0.9.0        
#  [4] htmlwidgets_1.6.4   ggrepel_0.9.6       lattice_0.22-6     
#  [7] tzdb_0.4.0          quadprog_1.5-8      vctrs_0.6.5        
# [10] tools_4.4.0         generics_0.1.3      parallel_4.4.0     
# [13] tibble_3.2.1        pkgconfig_2.0.3     data.table_1.17.0  
# [16] SQUAREM_2021.1      RcppParallel_5.1.10 lifecycle_1.0.4    
# [19] truncnorm_1.0-9     farver_2.1.2        compiler_4.4.0     
# [22] stringr_1.5.1       git2r_0.33.0        progress_1.2.3     
# [25] munsell_0.5.1       RhpcBLASctl_0.23-42 httpuv_1.6.15      
# [28] htmltools_0.5.8.1   sass_0.4.10         yaml_2.3.10        
# [31] lazyeval_0.2.2      plotly_4.10.4       crayon_1.5.3       
# [34] later_1.4.2         pillar_1.10.2       jquerylib_0.1.4    
# [37] whisker_0.4.1       tidyr_1.3.1         MASS_7.3-61        
# [40] uwot_0.2.3          cachem_1.1.0        gtools_3.9.5       
# [43] tidyselect_1.2.1    digest_0.6.37       Rtsne_0.17         
# [46] stringi_1.8.7       reshape2_1.4.4      purrr_1.0.4        
# [49] ashr_2.2-66         labeling_0.4.3      rprojroot_2.0.4    
# [52] fastmap_1.2.0       grid_4.4.0          colorspace_2.1-1   
# [55] cli_3.6.5           invgamma_1.1        magrittr_2.0.3     
# [58] withr_3.0.2         prettyunits_1.2.0   scales_1.3.0       
# [61] promises_1.3.2      rmarkdown_2.29      httr_1.4.7         
# [64] startupmsg_0.9.6.1  workflowr_1.7.1     hms_1.1.3          
# [67] pbapply_1.7-2       evaluate_1.0.3      distr_2.9.3        
# [70] knitr_1.50          viridisLite_0.4.2   irlba_2.3.5.1      
# [73] rlang_1.1.6         Rcpp_1.0.14         mixsqp_0.3-54      
# [76] glue_1.8.0          rstudioapi_0.16.0   jsonlite_2.0.0     
# [79] plyr_1.8.9          R6_2.6.1            sfsmisc_1.1-18     
# [82] fs_1.6.6