Last updated: 2023-08-15

Checks: 7 0

Knit directory: SuperCellCyto-analysis/

This reproducible R Markdown analysis was created with workflowr (version 1.7.0). 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: up-to-date

Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.

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(42)

The command set.seed(42) 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: 7baf7b7

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 7baf7b7. 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: 


Ignored files:
    Ignored:    .DS_Store
    Ignored:    .Rhistory
    Ignored:    .Rproj.user/
    Ignored:    analysis/.DS_Store
    Ignored:    code/.DS_Store
    Ignored:    code/b_cell_identification/.DS_Store
    Ignored:    code/batch_correction/.DS_Store
    Ignored:    code/explore_supercell_purity_clustering/.DS_Store
    Ignored:    code/explore_supercell_purity_clustering/functions/.DS_Store
    Ignored:    code/explore_supercell_purity_clustering/louvain_all_cells/.DS_Store
    Ignored:    code/label_transfer/.Rhistory
    Ignored:    data/.DS_Store
    Ignored:    data/bodenmiller_cytof/
    Ignored:    data/explore_supercell_purity_clustering/
    Ignored:    data/haas_bm/
    Ignored:    data/oetjen_bm_dataset/
    Ignored:    data/trussart_cytofruv/
    Ignored:    output/.DS_Store
    Ignored:    output/bodenmiller_cytof/
    Ignored:    output/explore_supercell_purity_clustering/
    Ignored:    output/label_transfer/
    Ignored:    output/oetjen_b_cell_panel/
    Ignored:    output/trussart_cytofruv/

Untracked files:
    Untracked:  code/b_cell_identification/cluster_singlecell.R
    Untracked:  code/b_cell_identification/rescale_supercell.R
    Untracked:  code/b_cell_identification/runtime_benchmark/
    Untracked:  code/batch_correction/benchmark_cycombine.R
    Untracked:  code/batch_correction/rescale_supercell.R
    Untracked:  code/batch_correction/runtime_benchmark/
    Untracked:  code/bodenmiller_data/benchmark_supercell_runtime.R
    Untracked:  code/compare_da_test_runtime.R
    Untracked:  code/krieg_melanoma/
    Untracked:  code/label_transfer/harmony_knn_singlecell.R
    Untracked:  code/label_transfer/seurat_rpca_singlecell.R
    Untracked:  output/krieg_melanoma/

Unstaged changes:
    Modified:   SuperCellCyto-analysis.Rproj
    Modified:   code/batch_correction/run_batch_correction_supercells.R
    Modified:   code/label_transfer/harmony_knn.R
    Modified:   code/label_transfer/seurat_rpca.R

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/run_time.Rmd) and HTML (docs/run_time.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
Rmd 7baf7b7 Givanna Putri 2023-08-15 wflow_publish("analysis/*")
html a55c3ba Givanna Putri 2023-07-28 Build site.
html 366514e Givanna Putri 2023-07-28 Build site.
Rmd 402358b Givanna Putri 2023-07-28 wflow_publish(c("analysis/*Rmd"))

Introduction

In this analysis, we examine the time required to create the supercells, as well as the run time improvement obtained by analysing supercells vs single cells.

library(here)
library(data.table)
library(stringr)
library(ggplot2)
library(scales)
library(viridis)

Function to convert the log 10 y-axis scale to actual number. Just so it’s easier to interpret. So instead of showing 1,2,3, which are the log 10 values, we show 10^1, 10^2, 10^3, etc.

convert_log10_to_actual <- function(labels, dp_below_1 = 2) {
  transformed_labs <- sapply(labels, function(x) {
    new_lab <- 10^x
    
    # if over 1, then round to integer, otherwise round to whatever dp_below_1 is
    if (new_lab >= 1) {
      new_lab <- round(new_lab)
    } else {
      new_lab <- round(new_lab, dp_below_1)
      new_lab <- sprintf(paste0("%.", dp_below_1, "f"), new_lab)
    }
    
    return(new_lab)
  })
}

Supercell creation

datasets <- c("samusik_all", "levine_32dim")
supercell_runtime <- lapply(datasets, function(dt_source) {
    dt <- fread(here("output", "explore_supercell_purity_clustering", "20230511", 
                            dt_source, "supercell_runs", "supercell_runtime.txt"),
                       sep = ":", header = FALSE, col.names = c("dataset", "duration_seconds"))
    dt$duration_seconds <- gsub(" sec elapsed", "", dt$duration_seconds)
    
    mean_duration <- mean(as.numeric(dt$duration_seconds))
    return(mean_duration)
})
names(supercell_runtime) <- str_to_title(datasets)
run_time_bcells <- fread(here("output", "oetjen_b_cell_panel", "20230511", "supercell_runtime.txt"),
                         sep = ":", header = FALSE, col.names = c("dataset", "duration_seconds"))
supercell_runtime[["Oetjen_bcells"]] <- mean(as.numeric(gsub(" sec elapsed", "", run_time_bcells$duration_seconds)))
run_time_trussart <- fread(here("output", "trussart_cytofruv", "20230515_supercell_out", "supercell_runtime.txt"),
                         sep = ":", header = FALSE, col.names = c("dataset", "duration_seconds"))
supercell_runtime[["Trussart_cytofruv"]] <- mean(as.numeric(gsub(" sec elapsed", "", run_time_trussart$duration_seconds)))
run_time_bodenmiller <- fread(here("output", "bodenmiller_cytof", "supercell_benchmark.txt"),
                         sep = ":", header = FALSE, col.names = c("dataset", "duration_seconds"))
supercell_runtime[["BCR_XL"]] <- mean(as.numeric(gsub(" sec elapsed", "", run_time_bodenmiller$duration_seconds)))
run_time_krieg <- fread(here("output", "krieg_melanoma", "supercell_benchmark.txt"),
                         sep = ":", header = FALSE, col.names = c("dataset", "duration_seconds"))
supercell_runtime[["Anti_PD1"]] <- mean(as.numeric(gsub(" sec elapsed", "", run_time_krieg$duration_seconds)))
supercell_runtime_dt <- data.table(
  dataset=names(supercell_runtime),
  duration_seconds=as.numeric(supercell_runtime)
)
supercell_runtime_dt[, duration_minutes := duration_seconds/60]
supercell_runtime_dt[, log10_duration_minutes := log10(duration_minutes)]

# manually entered this so watch out if it is wrong
supercell_runtime_dt[, ncells := c(841644, 265627, 8314260, 8589739, 172791, 85715)]

# mainly for plotting so we plot from smallest to largest
supercell_runtime_dt <- supercell_runtime_dt[order(ncells)]

supercell_runtime_dt[, ncells_pretty_label := prettyNum(ncells, big.mark=",", scientific=FALSE)]
supercell_runtime_dt[, ncells_pretty_label := factor(ncells_pretty_label, levels=supercell_runtime_dt$ncells_pretty_label)]
y_axis_labs <- seq(-1.5,2.5,0.5)
ggplot(supercell_runtime_dt, aes(x = ncells_pretty_label, y = log10_duration_minutes)) +
  geom_point(aes(colour = dataset), size = 2) +
  theme_classic() +
  scale_y_continuous(breaks = y_axis_labs, label = convert_log10_to_actual(y_axis_labs,1)) +
  labs(y = "Duration (minutes)", x = "Number of cells", color = "Dataset", title = "Time taken to create supercells") +
  theme(axis.text.x = element_text(angle = 45,  hjust=1), legend.position="bottom")

Version Author Date
366514e Givanna Putri 2023-07-28

Batch correction

cycombine_duration_run1 <- fread("output/trussart_cytofruv/benchmark_duration/cycombine_benchmark_run1.txt")
cycombine_duration_run1[, run := 'run_1']
cycombine_duration_run2 <- fread("output/trussart_cytofruv/benchmark_duration/cycombine_benchmark_run2.txt")
cycombine_duration_run2[, run := 'run_2']

cycombine_duration <- rbind(cycombine_duration_run1, cycombine_duration_run2)
cycombine_duration <- cycombine_duration[, c("V6","V4", "V7", "run")]
names(cycombine_duration) <- c("gamma", "setting", "duration_seconds", "run")

cycombine_duration[, gamma := gsub(":", "", gamma)]
cycombine_duration[, gamma := gsub("singlecell", "single cell", gamma)]
cycombine_duration[, gamma := gsub("gam", "", gamma)]

cycombine_duration <- cycombine_duration[, .(duration_seconds = mean(duration_seconds)), by=c('gamma', 'setting')]
cycombine_duration[, algorithm := 'cyCombine']
cytofruv_duration_singlecell <- lapply(seq(24), function(setting) {
    dt_run1 <- fread(paste0("output/trussart_cytofruv/benchmark_duration/cytofruv/run1/cytofruv_singlecell_setting", setting, ".txt"))
    dt_run2 <- fread(paste0("output/trussart_cytofruv/benchmark_duration/cytofruv/run2/cytofruv_singlecell_setting", setting, ".txt"))
    mean_duration <- mean(c(as.numeric(dt_run1$V5), as.numeric(dt_run2$V5)))
    return(mean_duration)
})
cytofruv_duration_singlecell_dt <- data.table(
  gamma = "single cell",
  setting = seq(24),
  duration_seconds = cytofruv_duration_singlecell
)

cytofruv_duration_supercell <- lapply(seq(10,40,10), function(gamma_val) {
  cytofruv_duration <- lapply(seq(24), function(setting) {
      dt_run1 <- fread(paste0("output/trussart_cytofruv/benchmark_duration/cytofruv/run1/cytofruv_supercell_gam", gamma_val, "_setting", setting, ".txt"))
      dt_run2 <- fread(paste0("output/trussart_cytofruv/benchmark_duration/cytofruv/run2/cytofruv_supercell_gam", gamma_val, "_setting", setting, ".txt"))
      mean_duration <- mean(c(as.numeric(dt_run1$V8), as.numeric(dt_run2$V8)))
      return(mean_duration)
  })
  cytofruv_duration_dt <- data.table(
    gamma = gamma_val,
    setting = seq(24),
    duration_seconds = cytofruv_duration
  )
  
  return(cytofruv_duration_dt)
})

cytofruv_duration_supercell[['singlecell']] <- cytofruv_duration_singlecell_dt
batch_correction_duration <- rbindlist(cytofruv_duration_supercell)
batch_correction_duration[, algorithm := "CytofRUV"]
batch_correction_duration <- rbind(batch_correction_duration, cycombine_duration)

batch_correction_duration[, algorithm := factor(algorithm)]
batch_correction_duration[, gamma := factor(gamma, levels = c("single cell", seq(5,50,5)))]

batch_correction_duration[, duration_seconds := as.numeric(duration_seconds)]
batch_correction_duration[, duration_minutes := duration_seconds/60]
batch_correction_duration[, log10_duration_minutes := log10(duration_minutes)]
y_axis_labs <- seq(-1.2,1.2,0.2)
ggplot(batch_correction_duration, aes(x = gamma, y = log10_duration_minutes, fill=algorithm)) +
  geom_boxplot(outlier.size=0.2, lwd=0.1) +
  scale_fill_brewer(palette="Dark2") +
  theme_classic() +
  facet_wrap(~algorithm) +
  theme(axis.text.x = element_text(angle = 45,  hjust=1), legend.position="none") +
  scale_y_continuous(breaks = y_axis_labs, label = convert_log10_to_actual(y_axis_labs)) +
  labs(y = "Duration (minutes)", x = "Gamma", title = "Time taken to do batch correction", fill='Algorithm')

Louvain clustering

louvain_runtime_supercell <- lapply(c("samusik_all", "levine_32dim"), function(dt_source) {
    dt <- lapply(seq(5, 50, by=5), function(gam) {
        dt <- fread(here("output", "explore_supercell_purity_clustering", "20230511", 
                         dt_source, "louvain_supercell_runs", paste0("gamma_", gam),
                         "louvain_supercell_runtime.txt"))
        
        res <- data.table(
          setting = factor(sapply(dt$V1, function(x) str_split_1(x, "_")[3])),
          duration_seconds = as.numeric(dt$V2)
        )
        
        med_dur <- res[, .(med_dur = mean(duration_seconds)), by = c("setting")]
        
        res <- data.table(
          gamma = gam,
          setting = med_dur$setting,
          duration_seconds = as.numeric(med_dur$med_dur),
          dataset = dt_source
        )
        return(res)
    })
    return(rbindlist(dt))
})
louvain_runtime_supercell <- rbindlist(louvain_runtime_supercell)

louvain_runtime_allcell <- lapply(c("samusik_all", "levine_32dim"), function(dt_source) {
    dt <- lapply(seq(10, 30, by=5), function(k) {
        dt <- fread(here("output", "explore_supercell_purity_clustering", "20230511", 
                         dt_source, "louvain_allcells", paste0("k", k),
                         "louvain_supercell_runtime.txt"))
        
        res <- data.table(
          gamma = "single cell",
          setting = sapply(dt$V1, function(x) str_split_1(x, "_")[3])[1],
          duration_seconds = mean(as.numeric(dt$V2)),
          dataset = dt_source
        )
        
        return(res)
    })
    return(rbindlist(dt))
})
louvain_runtime_allcell <- rbindlist(louvain_runtime_allcell)

louvain_runtime_others <- rbind(
  louvain_runtime_allcell, 
  louvain_runtime_supercell
)
k_vals <- c(3,5,10,15,20)
oetjen_duration_singlecell <- lapply(k_vals, function(setting) {
    dt <- fread(paste0("output/oetjen_b_cell_panel/benchmark_runtime/run1/duration_louvain_singlecell_k", setting, ".txt"))
    return(as.numeric(dt$V2))
})
oetjen_duration_singlecell_dt <- data.table(
  gamma = "single cell",
  setting = k_vals,
  duration_seconds = oetjen_duration_singlecell
)

oetjen_duration_supercell <- lapply(seq(10,40,10), function(gamma_val) {
  oetjen_duration <- lapply(k_vals, function(setting) {
      dt <- fread(paste0("output/oetjen_b_cell_panel/benchmark_runtime/run1/duration_louvain_supercell_gamma", gamma_val, "_k", setting, ".txt"))
      return(as.numeric(dt$V2))
  })
  oetjen_duration_dt <- data.table(
    gamma = gamma_val,
    setting = k_vals,
    duration_seconds = oetjen_duration
  )
  
  return(oetjen_duration_dt)
})

oetjen_duration_supercell[['singlecell']] <- oetjen_duration_singlecell_dt
oetjen_duration_dt <- rbindlist(oetjen_duration_supercell)
oetjen_duration_dt[, dataset := "oetjen_bcells"]
louvain_runtime <- rbind(louvain_runtime_others, oetjen_duration_dt)

louvain_runtime[, duration_seconds := as.numeric(duration_seconds)]
louvain_runtime[, log10_duration_seconds := log10(duration_seconds)]
louvain_runtime[, duration_minutes := duration_seconds/60]
louvain_runtime[, log10_duration_minutes := log10(duration_minutes)]
louvain_runtime[, gamma := factor(gamma, levels = c("single cell", seq(5,50,5)))]
y_axis_labs <- seq(-2,3.5,0.5)
ggplot(louvain_runtime, aes(x = gamma, y = log10_duration_minutes, fill=dataset)) +
  geom_boxplot(outlier.size = 0.1, lwd=0.1) +
  scale_fill_viridis(discrete = TRUE, alpha=0.6) +
  theme_classic() +
  facet_wrap(~dataset) +
  scale_y_continuous(breaks = y_axis_labs, label = convert_log10_to_actual(y_axis_labs)) +
  labs(y = "Duration (minutes)", x = "Gamma", title = "Time taken to run Louvain clustering", fill="Dataset") +
  theme(axis.text.x = element_text(angle = 45,  hjust=1), legend.position="none")

FlowSOM clustering

fsom_runtime_supercell <- lapply(c("samusik_all", "levine_32dim"), function(dt_source) {
    dt <- lapply(seq(5, 50, by=5), function(gam) {
        dt <- fread(here("output", "explore_supercell_purity_clustering", "20230511", 
                         dt_source, "flowsom_supercell_runs", paste0("gamma_", gam),
                         "flowsom_supercell_runtime.txt"))
        res <- data.table(
          setting = factor(sapply(dt$V1, function(x) paste0(str_split_1(x, "_")[3], "_", str_split_1(x, "_")[4]))),
          duration_seconds = as.numeric(dt$V2)
        )
        
        med_dur <- res[, .(med_dur = mean(duration_seconds)), by = c("setting")]
        
        res <- data.table(
          gamma = gam,
          setting = med_dur$setting,
          duration_seconds = as.numeric(med_dur$med_dur),
          dataset = dt_source
        )
        return(res)
    })
    return(rbindlist(dt))
})
fsom_runtime_supercell <- rbindlist(fsom_runtime_supercell)

fsom_runtime_allcell <- lapply(c("samusik_all", "levine_32dim"), function(dt_source) {
    dt <- fread(here("output", "explore_supercell_purity_clustering", "20230509", 
                         dt_source, "flowsom_allcells", "flowsom_allcell_runtime.txt"))
    res <- data.table(
      setting = factor(sapply(dt$V1, function(x) paste0(str_split_1(x, "_")[3], "_", str_split_1(x, "_")[4]))),
      duration_seconds = as.numeric(dt$V2)
    )
    
    med_dur <- res[, .(med_dur = mean(duration_seconds)), by = c("setting")]
    
    res <- data.table(
      gamma = "single cell",
      setting = med_dur$setting,
      duration_seconds = as.numeric(med_dur$med_dur),
      dataset = dt_source
    )
    
    return(res)
})
fsom_runtime_allcell <- rbindlist(fsom_runtime_allcell)

fsom_runtime <- rbind(
  fsom_runtime_allcell, 
  fsom_runtime_supercell
)

fsom_runtime[, duration_seconds := as.numeric(duration_seconds)]
fsom_runtime[, log10_duration_seconds := log10(duration_seconds)]

fsom_runtime[, gamma := factor(gamma, levels = c("single cell", seq(5,50,5)))]
ggplot(fsom_runtime, aes(x = gamma, y = duration_seconds, fill=dataset)) +
  scale_fill_viridis(discrete = TRUE, alpha=0.6) +
  geom_boxplot(outlier.size=0.2, lwd=0.2) +
  theme_classic() +
  facet_wrap(~dataset) +
  scale_y_continuous(breaks = pretty_breaks(n=20)) +
  labs(y = "Duration (seconds)", x = "Gamma", title = "Time taken to run FlowSOM clustering", fill="Dataset") +
  theme(axis.text.x = element_text(angle = 45,  hjust=1), legend.position="none")

Label transfer

dur_files <- c("harmony_knn_runtime_singlecell.txt", "harmony_knn_runtime.txt", 
               "seurat_rPCA_singlecell_runtime.txt", "seurat_rPCA_runtime.txt")
label_trans <- sapply(dur_files, function(f) {
  dt <- fread(paste0("output/label_transfer/", f))
  return(dt$V3)
})
label_trans_dt <- data.table(
  approah = names(label_trans),
  duration_seconds = label_trans,
  duration_minutes = label_trans/60
)
label_trans_dt
                              approah duration_seconds duration_minutes
1: harmony_knn_runtime_singlecell.txt          323.237         5.387283
2:            harmony_knn_runtime.txt           62.565         1.042750
3: seurat_rPCA_singlecell_runtime.txt         2197.163        36.619383
4:            seurat_rPCA_runtime.txt          575.482         9.591367

sessionInfo()
R version 4.2.3 (2023-03-15)
Platform: aarch64-apple-darwin20 (64-bit)
Running under: macOS Monterey 12.6

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/4.2-arm64/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.2-arm64/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

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

other attached packages:
[1] viridis_0.6.2     viridisLite_0.4.1 scales_1.2.1      ggplot2_3.4.1    
[5] stringr_1.5.0     data.table_1.14.8 here_1.0.1        workflowr_1.7.0  

loaded via a namespace (and not attached):
 [1] tidyselect_1.2.0   xfun_0.39          bslib_0.4.2        colorspace_2.1-0  
 [5] vctrs_0.5.2        generics_0.1.3     htmltools_0.5.4    yaml_2.3.7        
 [9] utf8_1.2.3         rlang_1.0.6        jquerylib_0.1.4    later_1.3.0       
[13] pillar_1.8.1       glue_1.6.2         withr_2.5.0        RColorBrewer_1.1-3
[17] lifecycle_1.0.3    munsell_0.5.0      gtable_0.3.1       evaluate_0.20     
[21] knitr_1.42         callr_3.7.3        fastmap_1.1.0      httpuv_1.6.9      
[25] ps_1.7.2           fansi_1.0.4        highr_0.10         Rcpp_1.0.10       
[29] promises_1.2.0.1   cachem_1.0.6       jsonlite_1.8.4     farver_2.1.1      
[33] fs_1.6.1           gridExtra_2.3      digest_0.6.31      stringi_1.7.12    
[37] processx_3.8.0     dplyr_1.1.0        getPass_0.2-2      rprojroot_2.0.3   
[41] grid_4.2.3         cli_3.6.0          tools_4.2.3        magrittr_2.0.3    
[45] sass_0.4.5         tibble_3.1.8       whisker_0.4.1      pkgconfig_2.0.3   
[49] rmarkdown_2.20     httr_1.4.4         rstudioapi_0.14    R6_2.5.1          
[53] git2r_0.31.0       compiler_4.2.3