Examine topic modeling results with k = 16 topics
Peter Carbonetto
Last updated: 2022-05-19
Checks: 5 2
Knit directory: lps/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.
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.
The global environment had objects present when the code in the R Markdown file was run. These objects 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. Use wflow_publish or wflow_build to ensure that the code is always run in an empty environment.
The following objects were defined in the global environment when these results were created:
| Name | Class | Size |
|---|---|---|
| counts | matrix | 40.9 Mb |
| dat | list | 77.6 Mb |
| de | topic_model_de_analysis;list | 48.7 Mb |
| de_merged | topic_model_de_analysis;list | 42.9 Mb |
| fit | poisson_nmf_fit;list | 15.8 Mb |
| fit_merged | multinom_topic_model_fit;list | 13.9 Mb |
| fit2 | multinom_topic_model_fit;list | 8 Mb |
| j | integer | 1.9 Mb |
| k | character | 112 bytes |
| p | gg;ggplot | 100.6 Kb |
| p1 | gg;ggplot | 111.6 Kb |
| p2 | gg;ggplot | 111.6 Kb |
| read_lps_data | function | 67.7 Kb |
| rows | integer | 1.5 Kb |
| samples | data.frame | 9.4 Kb |
| topic_colors | character | 320 bytes |
| topics | character | 432 bytes |
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.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.
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 219b260. 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: data/raw_read_counts.csv.gz
Ignored: output/fit-lps-k=16.RData
Untracked files:
Untracked: analysis/volcano_plot_k1+k5.html
Untracked: analysis/volcano_plot_k1.html
Untracked: analysis/volcano_plot_k10.html
Untracked: analysis/volcano_plot_k11.html
Untracked: analysis/volcano_plot_k12.html
Untracked: analysis/volcano_plot_k13.html
Untracked: analysis/volcano_plot_k14.html
Untracked: analysis/volcano_plot_k15.html
Untracked: analysis/volcano_plot_k16.html
Untracked: analysis/volcano_plot_k2+k13.html
Untracked: analysis/volcano_plot_k2.html
Untracked: analysis/volcano_plot_k3.html
Untracked: analysis/volcano_plot_k4.html
Untracked: analysis/volcano_plot_k5.html
Untracked: analysis/volcano_plot_k6+k14.html
Untracked: analysis/volcano_plot_k6.html
Untracked: analysis/volcano_plot_k7.html
Untracked: analysis/volcano_plot_k8.html
Untracked: analysis/volcano_plot_k9.html
Unstaged changes:
Modified: analysis/examine_topic_model_k16.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/examine_topic_model_k16.Rmd) and HTML (docs/examine_topic_model_k16.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 | 2e949c0 | Peter Carbonetto | 2022-05-10 | Added links to volcano plots. |
| Rmd | ba08036 | Peter Carbonetto | 2022-05-10 | workflowr::wflow_publish(“examine_topic_model_k16.Rmd”, verbose = TRUE) |
| html | 29047dc | Peter Carbonetto | 2022-05-10 | Build site. |
| Rmd | 23fcc2b | Peter Carbonetto | 2022-05-10 | workflowr::wflow_publish(“examine_topic_model_k16.Rmd”, verbose = TRUE) |
| html | 6e95930 | Peter Carbonetto | 2022-05-10 | Fixed interactive volcano_plots. |
| Rmd | a136de3 | Peter Carbonetto | 2022-05-10 | workflowr::wflow_publish(“examine_topic_model_k16.Rmd”, verbose = TRUE) |
| html | bef8deb | Peter Carbonetto | 2022-05-10 | Added volcano plots. |
| Rmd | 567870b | Peter Carbonetto | 2022-05-10 | workflowr::wflow_publish(“examine_topic_model_k16.Rmd”, verbose = TRUE) |
| html | 6dabc4d | Peter Carbonetto | 2022-05-05 | Added another structure plot to examine_topic_model_k16 analysis. |
| Rmd | 1899304 | Peter Carbonetto | 2022-05-05 | workflowr::wflow_publish(“examine_topic_model_k16.Rmd”) |
| html | b5a1eae | Peter Carbonetto | 2022-05-04 | Build site. |
| Rmd | a628c54 | Peter Carbonetto | 2022-05-04 | workflowr::wflow_publish(“examine_topic_model_k16.Rmd”, verbose = TRUE) |
| Rmd | e01a993 | Peter Carbonetto | 2022-05-04 | First build of overview page. |
Add text here giving an overview of this analysis.
Load the packages used in the analysis.
library(data.table)
library(fastTopics)
library(ggplot2)
library(cowplot)
source("../code/lps_data.R")Initialize the sequence of pseudorandom numbers.
set.seed(1)Load the count data.
dat <- read_lps_data("../data/raw_read_counts.csv.gz")
samples <- dat$samples
counts <- dat$countsRemove genes with very low (or no) expression.
j <- which(colSums(counts) > 20)
counts <- counts[,j]Load the results of the topic modeling analysis.
load("../output/fit-lps-k=16.RData")Plot the improvement in the solution over time.
p1 <- plot_progress(fit,x = "timing",y = "loglik",colors = "black",
add.point.every = 10,e = 1e-4) +
guides(color = "none",fill = "none",shape = "none",
linetype = "none",size = "none")
p2 <- plot_progress(fit,x = "timing",y = "res",colors = "black",
add.point.every = 10,e = 1e-4) +
guides(color = "none",fill = "none",shape = "none",
linetype = "none",size = "none")
plot_grid(p1,p2)
Visualize the structure identified in each of the tissues using a Structure plot, in which the samples in each tissue are ordered by time in which the sample was taken:
set.seed(1)
rows <- order(samples$timepoint)
topic_colors <- c("darkblue","dodgerblue","darkorange","forestgreen",
"limegreen","tomato","darkred","olivedrab","magenta",
"darkmagenta","sienna","royalblue","lightskyblue",
"gold","red","cyan")
p <- structure_plot(fit,grouping = samples$tissue,gap = 5,
colors = topic_colors,
topics = c(15,3,4,5,6,7,8,9,10,11,12,13,2,14,1,16),
loadings_order = rows) +
theme(legend.key.height = unit(0.15,"cm"),
legend.text = element_text(size = 7))
print(p)
| Version | Author | Date |
|---|---|---|
| 6dabc4d | Peter Carbonetto | 2022-05-05 |
There is a single topic (topic 9, blue in the plot below) that is capturing changes in expression over time across many tissues. Two other topics (topics 1 and 6) show similar patterns, except these patterns are specific to two tissues (PBMC and LI).
set.seed(1)
topic_colors <- c("gold","darkorange","dodgerblue","white")
fit2 <- poisson2multinom(fit)
fit2 <- merge_topics(fit2,paste0("k",setdiff(1:16,c(1,6,9))))
colnames(fit2$L) <- c("k1","k6","k9","other")
p <- structure_plot(fit2,grouping = samples$tissue,gap = 5,
colors = topic_colors,topics = c(4,1:3),
loadings_order = rows)
print(p)
| Version | Author | Date |
|---|---|---|
| 6dabc4d | Peter Carbonetto | 2022-05-05 |
These volcano plots summarize the results of the DE analysis for topics capturing different tissues (and topic 9, which is capturing changes in expression at different time points):
topics <- colnames(de_merged$z)
p <- vector("list",13)
names(p) <- topics
for (k in topics) {
p[[k]] <- volcano_plot(de_merged,k = k,ymax = 500) +
guides(fill = "none")
volcano_plotly(de_merged,k = k,ymax = 500,
file = paste("volcano_plot_",k,".html",sep = ""))
}
do.call("plot_grid",c(p,list(ncol = 3,nrow = 5)))
These volcano plots summarize the results of the DE analysis for topics capturing expression patterns within tissues:
topics <- c("k1","k2","k5","k6","k13","k14")
p <- vector("list",6)
names(p) <- topics
for (k in topics) {
p[[k]] <- volcano_plot(de,k = k,ymax = 300) +
guides(fill = "none")
volcano_plotly(de,k = k,ymax = 300,
file = paste("volcano_plot_",k,".html",sep = ""))
}
do.call("plot_grid",c(p,list(ncol = 3,nrow = 2)))
These results may also be browsed interactively: k1, k2, k3, k4, k5, k6, k7, k8, k9, k10, k11, k12, k13, k14, k15, k16, k1+k5, k2+k13, k6+k14.
sessionInfo()
# R version 3.6.2 (2019-12-12)
# Platform: x86_64-apple-darwin15.6.0 (64-bit)
# Running under: macOS Catalina 10.15.7
#
# Matrix products: default
# BLAS: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRblas.0.dylib
# LAPACK: /Library/Frameworks/R.framework/Versions/3.6/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] cowplot_1.0.0 ggplot2_3.3.5 fastTopics_0.6-131 data.table_1.12.8
#
# loaded via a namespace (and not attached):
# [1] mcmc_0.9-6 fs_1.5.2 progress_1.2.2 httr_1.4.2
# [5] rprojroot_1.3-2 tools_3.6.2 backports_1.1.5 bslib_0.3.1
# [9] utf8_1.1.4 R6_2.4.1 irlba_2.3.3 uwot_0.1.10
# [13] DBI_1.1.0 lazyeval_0.2.2 colorspace_1.4-1 withr_2.5.0
# [17] tidyselect_1.1.1 prettyunits_1.1.1 compiler_3.6.2 git2r_0.29.0
# [21] quantreg_5.54 SparseM_1.78 plotly_4.9.2 labeling_0.3
# [25] sass_0.4.0 scales_1.1.0 SQUAREM_2017.10-1 quadprog_1.5-8
# [29] pbapply_1.5-1 mixsqp_0.3-46 stringr_1.4.0 digest_0.6.23
# [33] rmarkdown_2.11 R.utils_2.11.0 MCMCpack_1.4-5 pkgconfig_2.0.3
# [37] htmltools_0.5.2 highr_0.8 fastmap_1.1.0 invgamma_1.1
# [41] htmlwidgets_1.5.1 rlang_0.4.11 rstudioapi_0.13 shiny_1.4.0
# [45] jquerylib_0.1.4 generics_0.0.2 farver_2.0.1 jsonlite_1.7.2
# [49] crosstalk_1.0.0 dplyr_1.0.7 R.oo_1.24.0 magrittr_2.0.1
# [53] Matrix_1.2-18 Rcpp_1.0.7 munsell_0.5.0 fansi_0.4.0
# [57] lifecycle_1.0.0 R.methodsS3_1.8.1 stringi_1.4.3 whisker_0.4
# [61] yaml_2.2.0 MASS_7.3-51.4 Rtsne_0.15 grid_3.6.2
# [65] parallel_3.6.2 promises_1.1.0 ggrepel_0.9.1 crayon_1.4.1
# [69] lattice_0.20-38 hms_1.1.0 knitr_1.37 pillar_1.6.2
# [73] glue_1.4.2 evaluate_0.14 RcppParallel_5.1.5 vctrs_0.3.8
# [77] httpuv_1.5.2 MatrixModels_0.4-1 gtable_0.3.0 purrr_0.3.4
# [81] tidyr_1.1.3 assertthat_0.2.1 ashr_2.2-54 xfun_0.29
# [85] mime_0.8 xtable_1.8-4 coda_0.19-3 later_1.0.0
# [89] viridisLite_0.3.0 truncnorm_1.0-8 tibble_3.1.3 workflowr_1.7.0
# [93] ellipsis_0.3.2