Deconvolution Benchmark in Human DLPFC
15 April 2025
Package
DeconvoBuddies 1.0.0
1 Introduction
1.1 What is Deconvolution?
Inferring the composition of different cell types in a bulk RNA-seq data
Deconvolution is a analysis that aims to calculate the proportion of different cell types that make up a sample of bulk RNA-seq, based off of cell type gene expression profiles in a single cell/nuclei RNA-seq dataset.
Figure 1. Overview on how to use single cell data to infer the composition of bulk RNA-seq samples
1.1.1 Deconvolution Methods
There are 20+ published reference based deconvolution methods. Below are a selection of 6 methods we tested in our deconvolution benchmark study.
| Approach | Method | Citation | Availability |
|---|---|---|---|
| weighted least squares | DWLS | Tsoucas et al, Nature Comm, 2019 | R Package CRAN |
| Bias correction: Assay | Bisque | Jew et al, Nature Comm, 2020 | R Package GitHub |
| Bias correction: Source | MuSiC | Wang et al, Nature Communications, 2019 | R Package GitHub |
| Machine Learning | CIBERSORTx | Newman et al., Nature BioTech, 2019 | Webtool |
| Bayesian | BayesPrism | Chu et al., Nature Cancer, 2022 | Webtool/R Package |
| linear | Hspe | Hunt et al., Ann. Appl. Stat, 2021 | R package GitHub |
1.2 Goals of this Vignette
We will be demonstrating how to use DeconvoBuddies tools when applying
deconvolution with the Bisque package.
- Install and load required packages
- Download DLPFC RNA-seq data, and reference snRNA-seq data
- Find marker genes with
DeconvoBuddiestools - Run deconvolution with
BisqueRNA - Explore deconvolution output and create composition plots with
DeconvoBuddiestools - Check proportion against RNAScope estimated proportions
1.3 Video Tutorial
Linked is a video from a presentation of an earlier version of this tutorial from our LIBD Rstats club.
2 Basics
3 1. Install and load required packages
R is an open-source statistical environment which can be easily
modified to enhance its functionality via packages.
DeconvoBuddies is a R package available via the
Bioconductor repository for packages. R can
be installed on any operating system from
CRAN after which you can install
DeconvoBuddies by using the following commands in your
R session:
3.1 Install DeconvoBuddies
if (!requireNamespace("BiocManager", quietly = TRUE)) {
install.packages("BiocManager")
}
BiocManager::install("DeconvoBuddies")
## Check that you have a valid Bioconductor installation
BiocManager::valid()3.2 Load Other Packages
Let’s load the packages we’ll be using in this vignette.
## Packages for different types of RNA-seq data structures in R
library("SummarizedExperiment")
library("SingleCellExperiment")
library("Biobase")
## For downloading data
library("spatialLIBD")
## For running deconvolution
library("BisqueRNA")
## Other helper packages for this vignette
library("dplyr")
library("tidyr")
library("tibble")
library("ggplot2")
## Our main package
library("DeconvoBuddies")4 2. Download DLPFC RNA-seq data, and reference snRNA-seq data.
4.1 Bulk RNA-seq data
Access the 110 sample Human DLPFC bulk RNA-seq dataset for LIBD described in more detail here. These
samples
are from 19 tissue blocks, and 10 neurotypical adult donors. Samples
were sequenced with two different library_types (polyA and
RiboZeroGold), and three different RNA_extraction (Cyto, Total, Nuc). There are in total n=110 samples after quality control.
## use fetch deconvo data to load rse_gene
rse_gene <- fetch_deconvo_data("rse_gene")
#> 2025-04-15 18:04:47.864327 Access ExperimentHub EH9625
#> see ?DeconvoBuddies and browseVignettes('DeconvoBuddies') for documentation
#> loading from cache
rse_gene
#> class: RangedSummarizedExperiment
#> dim: 21745 110
#> metadata(1): SPEAQeasy_settings
#> assays(2): counts logcounts
#> rownames(21745): ENSG00000227232.5 ENSG00000278267.1 ... ENSG00000210195.2 ENSG00000210196.2
#> rowData names(11): Length gencodeID ... gencodeTx passExprsCut
#> colnames(110): 2107UNHS-0291_Br2720_Mid_Bulk 2107UNHS-0291_Br2720_Mid_Cyto ... AN00000906_Br8667_Mid_Cyto
#> AN00000906_Br8667_Mid_Nuc
#> colData names(80): SAMPLE_ID Sample ... diagnosis qc_class
# lobstr::obj_size(rse_gene)
# 41.16 MB
## Use gene "Symbol" as identifiers for the genes in rownames(rse_gene)
rownames(rse_gene) <- rowData(rse_gene)$Symbol
## bulk RNA seq samples were sequenced with different library types,
## and RNA extractions
table(rse_gene$library_type, rse_gene$library_prep)
#>
#> Bulk Cyto Nuc
#> RiboZeroGold 19 19 17
#> polyA 19 18 184.2 Reference snRNA-seq data
This data is paired with a single nucleus RNA-seq data set from
spatialLIBD. This dataset can be accessed with
spatialLIBD::fetch_data().
## Use spatialLIBD to fetch the snRNA-seq dataset used in this project
sce_path_zip <- fetch_deconvo_data("sce")
#> 2025-04-15 18:04:52.242972 loading file /home/biocbuild/.cache/R/BiocFileCache/1bacc542b1b486_sce_DLPFC_annotated.zip%3Fdl%3D1
## unzip and load the data
sce_path <- unzip(sce_path_zip, exdir = tempdir())
sce <- HDF5Array::loadHDF5SummarizedExperiment(
file.path(tempdir(), "sce_DLPFC_annotated")
)
# lobstr::obj_size(sce)
# 172.28 MB
## exclude Ambiguous cell type
sce <- sce[, sce$cellType_broad_hc != "Ambiguous"]
sce$cellType_broad_hc <- droplevels(sce$cellType_broad_hc)
## Check the number of genes by the number of nuclei that we
## have to work with:
dim(sce)
#> [1] 36601 56447
## Check the broad cell type distribution
table(sce$cellType_broad_hc)
#>
#> Astro EndoMural Micro Oligo OPC Excit Inhib
#> 3979 2157 1601 10894 1940 24809 11067
## We're going to subset to the first 5k genes to save memory
## just for this example. You wouldn't do this on a full analysis.
sce <- sce[seq_len(5000), ]4.3 Orthogonal Cell Type Proportion from RNAScope/IF
An alternative method for calculating cell type proportions is through imaging a slice of tissue with cell type probes using single molecule fluorescent in situ hybridization (smFISH) experiments performed with RNAScope/IF (ImmunoFluorescence). Then analyse the image to count cells and annotate them by cell type. In this study we used HALO from Indica Labs for this step.
The cell type proportions from the RNAScope/IF experiment will be used to evaluate the accuracy of cell type proportion estimates.
Figure 2. RNAScope/IF measures the cell type proportions through imaging
The RNAScope/IF proportion data is stored as a data.frame object in
DeconvoBuddies::RNAScope_prop.
Key columns in RNAScope_prop:
SAMPLE_ID: DLPFC Tissue block + RNAScope combination.Sample: DLFPC Tissue block (Donor BrNum + DLPFC position).cell_type: The cell type measured.n_cell: the number of cells counted for the Sample and cell type.prop: the calculated cell type proportion from n_cell
# Access the RNAScope proportion data.frame
data("RNAScope_prop")
head(RNAScope_prop)
#> SAMPLE_ID Sample Combo cell_type Confidence n_cell prop n_cell_sn prop_sn
#> 1 Br2720P_CIRCLE Br2720_post Circle Astro High 3385 0.09357808 81 0.04287983
#> 2 Br2720P_CIRCLE Br2720_post Circle EndoMural High 1680 0.04644348 66 0.03493912
#> 3 Br2720P_CIRCLE Br2720_post Circle Inhib High 3092 0.08547812 94 0.04976178
#> 4 Br2720P_CIRCLE Br2720_post Circle Other High 28016 0.77450032 NA NA
#> 5 Br2720P_STAR Br2720_post Star Excit High 8687 0.23132639 1198 0.63419799
#> 6 Br2720P_STAR Br2720_post Star Micro High 1416 0.03770671 47 0.02488089
## plot the RNAScope compositions
plot_composition_bar(
prop_long = RNAScope_prop,
sample_col = "SAMPLE_ID",
x_col = "SAMPLE_ID",
add_text = FALSE
) +
facet_wrap(~Combo, nrow = 2, scales = "free_x")
Above we can see the cell proportions for each of the samples in either the Circle or the Star combination of RNAScope/IF probes. Each combination of RNAScope/IF probes was able to assess different sets of cell types. Two of them had to be used to measure the number of cell types studied in this case as there are limits on how many unique probes you can measure in a given RNAScope/IF experiment. For more details about the RNAScope/IF combinations, check the paper describing this study.
5 3. Select Marker Genes
Marker genes are genes with high expression in one cell type and low expression in other cell types, or “cell-type specific” expression. These genes can be used to learn more about the identity and function of cell types, but here we are interested in using a sets of cell type specific marker genes to reduce noise in deconvolution and increase accuracy.
We have developed a method for finding marker genes called the “Mean Ratio”. We calculate the MeanRatio for a target cell type for each gene by dividing the mean expression of the target cell by the mean expression of the next highest non-target cell type. Genes with the highest MeanRatio values are selected as marker genes.
For a tutorial on marker gene selection check out Vignette: Finding Marker Genes with Deconvo Buddies.
Figure 3. Mean Ratio calculation process compared to 1vALL Marker Gene selection
5.1 Use get_mean_ratio() to find marker genes.
The function DeconvoBuddies::get_mean_ratio() calculates the
MeanRatio and the rank of genes for a specified cell type annotation
in an SingleCellExperiment object.
# calculate the Mean Ratio of genes for each cell type
marker_stats <- get_mean_ratio(sce,
cellType_col = "cellType_broad_hc",
gene_ensembl = "gene_id",
gene_name = "gene_name"
)
# check the top gene ranked gene for each cell type
marker_stats |>
group_by(cellType.target) |>
slice(1)
#> # A tibble: 7 × 10
#> # Groups: cellType.target [7]
#> gene cellType.target mean.target cellType.2nd mean.2nd MeanRatio MeanRatio.rank MeanRatio.anno gene_ensembl gene_name
#> <chr> <fct> <dbl> <fct> <dbl> <dbl> <int> <chr> <chr> <chr>
#> 1 PRDM… Astro 1.97 EndoMural 0.142 13.9 1 Astro/EndoMur… ENSG0000014… PRDM16
#> 2 SLC2… EndoMural 1.49 Astro 0.146 10.2 1 EndoMural/Ast… ENSG0000011… SLC2A1
#> 3 LINC… Micro 1.57 Excit 0.0640 24.5 1 Micro/Excit: … ENSG0000023… LINC01141
#> 4 AC01… Oligo 2.37 OPC 0.147 16.1 1 Oligo/OPC: 16… ENSG0000022… AC012494…
#> 5 MIR3… OPC 1.58 Excit 0.205 7.69 1 OPC/Excit: 7.… ENSG0000022… MIR3681HG
#> 6 AC01… Excit 1.01 Inhib 0.135 7.51 1 Excit/Inhib: … ENSG0000023… AC011995…
#> 7 LYPD… Inhib 1.20 Excit 0.0967 12.4 1 Inhib/Excit: … ENSG0000015… LYPD6BThe columns of this table are documented in detail at DeconvoBuddies::get_mean_ratio(). Though cellType.target lists the target cell type for which the MeanRatio is being calculated. It is the numerator of the MeanRatio. The cellType.2nd lists the cell type against which the target cell type is being compared. It is the denominator of the MeanRatio.
5.2 Plot the top marker genes
Use DeconvoBuddies plotting tools to quickly plot the gene expression
of the top 4 Excitatory neuron marker genes across the
cellType_broad_hc cell type annotations.
# plot expression across cell type the top 4 Excit marker genes
plot_marker_express(sce,
stats = marker_stats,
cell_type = "Excit",
cellType_col = "cellType_broad_hc",
rank_col = "MeanRatio.rank",
anno_col = "MeanRatio.anno",
gene_col = "gene"
)
Looks nice and cell type specific!
Note how plot_marker_express() lists the 2 cell types that are being compared that result in the specific numerical MeanRatio value being displayed. As we specified that our target cell type is "Excit" in our call to plot_marker_express(), the numerator is Excit in all the panels shown above.
5.2.1 Create a List of Marker Genes
With the MeanRatio calculated, we will select the top 25 highest Mean
Ratio genes for each cell type, that also exists in the bulk data
rse_gene.
# select top 25 marker genes for each cell type, that are also in rse_gene
marker_genes <- marker_stats |>
filter(MeanRatio.rank <= 25 & gene %in% rownames(rse_gene))
# check how many genes for each cell type (some genes are not in both datasets)
marker_genes |> count(cellType.target)
#> # A tibble: 7 × 2
#> cellType.target n
#> <fct> <int>
#> 1 Astro 22
#> 2 EndoMural 25
#> 3 Micro 24
#> 4 Oligo 23
#> 5 OPC 22
#> 6 Excit 21
#> 7 Inhib 24
# create a vector of marker genes to subset data before deconvolution
marker_genes <- marker_genes |> pull(gene)6 4. Prep Data and Run Bisque
Bisque is an R package for cell
type deconvolution. In our deconvolution benchmark, we found it was a
top performing method. Below we will briefly show how to run Bisque’s
“reference based decomposition (deconvolution)”.
6.1 Prepare data
To run Bisque the snRNA-seq and bulk data must first be converted to
Biobase::ExpressionSet() format. We will subset our data to our selected
MeanRatio marker genes.
The snRNA-seq data must also be filtered for cells with no counts across marker genes.
## convert bulk data to Expression set, sub-setting to marker genes
## include sample ID
exp_set_bulk <- Biobase::ExpressionSet(
assayData = assays(rse_gene[marker_genes, ])$counts,
phenoData = AnnotatedDataFrame(
as.data.frame(colData(rse_gene))[c("SAMPLE_ID")]
)
)
## convert snRNA-seq data to Expression set, sub-setting to marker genes
## include cell type and donor information
exp_set_sce <- Biobase::ExpressionSet(
assayData = as.matrix(assays(sce[marker_genes, ])$counts),
phenoData = AnnotatedDataFrame(
as.data.frame(colData(sce))[, c("cellType_broad_hc", "BrNum")]
)
)
## check for nuclei with 0 marker expression
zero_cell_filter <- colSums(exprs(exp_set_sce)) != 0
message("Exclude ", sum(!zero_cell_filter), " cells")
#> Exclude 0 cells
exp_set_sce <- exp_set_sce[, zero_cell_filter]6.2 Run Bisque
Bisque needs the bulk and single cell ExpressionSet we prepared
above, plus columns in the single cell data that specify the cell type
annotation to use cellType_broad_hc and donor id (BrNum in this
data).
## Run Bisque with bulk and single cell ExpressionSet inputs
est_prop <- ReferenceBasedDecomposition(
bulk.eset = exp_set_bulk,
sc.eset = exp_set_sce,
cell.types = "cellType_broad_hc",
subject.names = "BrNum",
use.overlap = FALSE
)
#> Decomposing into 7 cell types.
#> Using 161 genes in both bulk and single-cell expression.
#> Converting single-cell counts to CPM and filtering zero variance genes.
#> Filtered 0 zero variance genes.
#> Converting bulk counts to CPM and filtering unexpressed genes.
#> Filtered 0 unexpressed genes.
#> Generating single-cell based reference from 56447 cells.
#> Inferring bulk transformation from single-cell alone.
#> Applying transformation to bulk samples and decomposing.6.3 Explore Output
Bisque predicts the proportion of the cell types in cellType_broad_hc
for each sample in the bulk data.
## Examine the output from Bisque, transpose to make it easier to work with
est_prop$bulk.props <- t(est_prop$bulk.props)
## sample x cell type matrix
head(est_prop$bulk.props)
#> Astro EndoMural Micro Oligo OPC Excit Inhib
#> 2107UNHS-0291_Br2720_Mid_Bulk 0.05786828 0.02373868 0.03522481 0.1885803 0.04396245 0.4976930 0.1529325
#> 2107UNHS-0291_Br2720_Mid_Cyto 0.09613503 0.01576541 0.04236801 0.2101065 0.01578266 0.4728291 0.1470133
#> 2107UNHS-0291_Br2720_Mid_Nuc 0.04995672 0.03686927 0.03709439 0.2216284 0.01988676 0.4494565 0.1851079
#> 2107UNHS-0291_Br6432_Ant_Bulk 0.06862682 0.03756306 0.03297615 0.1937775 0.04199513 0.4474728 0.1775886
#> 2107UNHS-0291_Br6432_Ant_Cyto 0.09860176 0.03144221 0.04428703 0.2241753 0.02038414 0.4445342 0.1365754
#> 2107UNHS-0291_Br6432_Ant_Nuc 0.05886274 0.04694488 0.03311716 0.2245202 0.02330603 0.4278448 0.18540427 5. Explore deconvolution output and create composition plots with DeconvoBuddies tools
To visualize the cell type proportion predictions, we can plot cell type
composition bar plots with DeconvoBuddies::plot_composition_bar(),
either the prediction for each sample, or the average proportion over a
group of samples.
## add Phenotype data to proportion estimates
pd <- colData(rse_gene) |>
as.data.frame() |>
select(SAMPLE_ID, Sample, library_combo)
## make proportion estimates long so they are ggplot2 friendly
prop_long <- est_prop$bulk.props |>
as.data.frame() |>
tibble::rownames_to_column("SAMPLE_ID") |>
tidyr::pivot_longer(!SAMPLE_ID, names_to = "cell_type", values_to = "prop") |>
left_join(pd)
#> Joining with `by = join_by(SAMPLE_ID)`
## create composition bar plots
## for all library preparations by sample n=110
## Remove the SAMPLE_ID names since they are very long using ggplot2::theme()
plot_composition_bar(
prop_long = prop_long,
sample_col = "SAMPLE_ID",
x_col = "SAMPLE_ID",
add_text = FALSE
) +
theme(axis.text.x = element_blank(), axis.ticks.x = element_blank())
## Average by brain donor
plot_composition_bar(
prop_long = prop_long,
sample_col = "SAMPLE_ID",
x_col = "Sample",
add_text = FALSE
)
## Each brain donor has up to 6 unique RNA library type and RNA extraction
## combinations
table(prop_long$Sample) / length(unique(prop_long$cell_type))
#>
#> Br2720_mid Br2720_post Br2743_ant Br3942_ant Br3942_mid Br6423_ant Br6423_post Br6432_ant Br6432_mid Br6471_ant
#> 5 6 5 6 6 6 6 6 6 6
#> Br6471_mid Br6522_mid Br6522_post Br8325_ant Br8325_mid Br8492_mid Br8492_post Br8667_ant Br8667_mid
#> 6 6 6 6 5 5 6 6 6
## Here are the 6 "SAMPLE_ID" values for brain donor with ID "Br8667_mid"
unique(prop_long$SAMPLE_ID[prop_long$Sample == "Br8667_mid"])
#> [1] "AN00000904_Br8667_Mid_Bulk" "AN00000904_Br8667_Mid_Cyto" "AN00000904_Br8667_Mid_Nuc" "AN00000906_Br8667_Mid_Bulk"
#> [5] "AN00000906_Br8667_Mid_Cyto" "AN00000906_Br8667_Mid_Nuc"This is a more complex scenario than the one from the introductory vignette where we were using random data. The first plot shows each of the n=110 bulk RNA-seq samples we have. The second plot shows the composition using the average across the 6 RNA library types and RNA extractions for each brain donor.
8 6. Check proportion against RNAScope/IF estimated proportions
Note to compare the deconvolution results to the RNAScope/IF proportions, Oligo and OPC need to be added together.
## Combine Oligo and OPC into OligoOPC
prop_long_opc <- prop_long |>
mutate(cell_type = gsub("Oligo|OPC", "OligoOPC", cell_type)) |>
group_by(SAMPLE_ID, Sample, library_combo, cell_type) |>
summarize(prop = sum(prop)) |>
ungroup()
#> `summarise()` has grouped output by 'SAMPLE_ID', 'Sample', 'library_combo'. You can override using the `.groups`
#> argument.
prop_long_opc |> count(cell_type)
#> # A tibble: 6 × 2
#> cell_type n
#> <chr> <int>
#> 1 Astro 110
#> 2 EndoMural 110
#> 3 Excit 110
#> 4 Inhib 110
#> 5 Micro 110
#> 6 OligoOPC 110
## Join RNAScope/IF and Bisque cell type proportions
prop_compare <- prop_long_opc |>
inner_join(
RNAScope_prop |>
select(Sample, cell_type, prop_RNAScope = prop, prop_sn),
by = c("Sample", "cell_type")
)We can now calculate the correlation plot a scatter plot of the proportions. Note that you can change the type of correlation algorithm used in the cor() function. The default method is "pearson".
## compute correlation with RNAScope/IF proportions
cor(prop_compare$prop, prop_compare$prop_RNAScope)
#> [1] 0.5357834
## Scatter plot with RNAScope/IF proportions
prop_compare |>
ggplot(aes(x = prop_RNAScope, y = prop, color = cell_type, shape = library_combo)) +
geom_point() +
geom_abline()
## correlation with snRNA-seq proportion
cor(prop_compare$prop, prop_compare$prop_sn)
#> [1] 0.8397746
## Scatter plot with RNAScope/IF proportions
prop_compare |>
ggplot(aes(x = prop_sn, y = prop, color = cell_type, shape = library_combo)) +
geom_point() +
geom_abline()
In the first plot we can see how Bisque overestimates the proportion of excitatory neurons (Excit) when compared against RNAScope/IF as all Excit points are higher than the diagonal line shown in black.
In the second plot we can see how Bisque proportions are closer to the snRNA-seq proportions. After all, given the challenges in generating orthogonal cell type proportion data, Bisque and other deconvolution methods were developed by comparing against simulated data, sometimes using pseudo-bulk sc/snRNA-seq data.
9 7. How to run deconvolution with hspe
hspe (formerly called dtangle) is another R package for deconvolution. It was also a top performing method in our deconvolution benchmark.
hspe is downloadable from GitHub but
can’t be shown on this vignette as Bioconductor packages cannot use packages from GitHub.
Below we show some example code to prepare input data and run hspe (not run
here):
if (FALSE) {
## Install hspe
# if (!requireNamespace("hspe", quietly = TRUE)) {
# ## Install version 0.1 which is the one listed on the main documentation
# ## at https://github.com/gjhunt/hspe/tree/main?tab=readme-ov-file#software
# remotes::install_url("https://github.com/gjhunt/hspe/raw/main/hspe_0.1.tar.gz")
# ## Alternatively, install from the latest version on GitHub with:
# # remotes::install_github("gjhunt/hspe", subdir = "lib_hspe")
#
# ## As of 2024-08-23, it's been 3 years since files were last modified
# ## at https://github.com/gjhunt/hspe/tree/main/lib_hspe.
# }
# library("hspe")
# pseudobulk the sce data by sample + cell type
sce_pb <- spatialLIBD::registration_pseudobulk(sce,
var_registration = "cellType_broad_hc",
var_sample_id = "Sample"
)
## extract the gene expression from the bulk rse_gene
mixture_samples <- t(assays(rse_gene)$logcounts)
mixture_samples[1:5, 1:5]
## create a vector of indexes of the different cell types
pure_samples <- rafalib::splitit(sce_pb$cellType_broad_hc)
## extract the the pseudobulked logcounts
reference_samples <- t(assays(sce_pb)$logcounts)
reference_samples[1:5, 1:5]
## check the number of genes match in the bulk (mixture) and single cell (reference)
ncol(mixture_samples) == ncol(reference_samples)
## run hspe
est_prop_hspe <- hspe(
Y = mixture_samples,
reference = reference_samples,
pure_samples = pure_samples,
markers = marker_genes,
seed = 10524
)
}10 Conclusion
In this vignette we have demonstrated some of the functions and data in
the DeconvoBuddies package, and how to use them in a deconvolution
workflow to predict cell type proportions. We used real data from a study
on human brain.
11 Reproducibility
The DeconvoBuddies package (Huuki-Myers, Maynard, Hicks et al., 2025) was made possible thanks to:
- R (R Core Team, 2025)
- BiocStyle (Oleś, 2025)
- knitr (Xie, 2025)
- RefManageR (McLean, 2017)
- rmarkdown (Allaire, Xie, Dervieux et al., 2024)
- sessioninfo (Wickham, Chang, Flight, Müller, and Hester, 2025)
- testthat (Wickham, 2011)
This package was developed using biocthis.
R session information.
#> ─ Session info ───────────────────────────────────────────────────────────────────────────────────────────────────────
#> setting value
#> version R version 4.5.0 RC (2025-04-04 r88126)
#> os Ubuntu 24.04.2 LTS
#> system x86_64, linux-gnu
#> ui X11
#> language (EN)
#> collate C
#> ctype en_US.UTF-8
#> tz America/New_York
#> date 2025-04-15
#> pandoc 2.7.3 @ /usr/bin/ (via rmarkdown)
#> quarto 1.5.57 @ /usr/local/bin/quarto
#>
#> ─ Packages ───────────────────────────────────────────────────────────────────────────────────────────────────────────
#> package * version date (UTC) lib source
#> abind 1.4-8 2024-09-12 [2] CRAN (R 4.5.0)
#> AnnotationDbi 1.70.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> AnnotationHub 3.16.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> attempt 0.3.1 2020-05-03 [2] CRAN (R 4.5.0)
#> backports 1.5.0 2024-05-23 [2] CRAN (R 4.5.0)
#> beachmat 2.24.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> beeswarm 0.4.0 2021-06-01 [2] CRAN (R 4.5.0)
#> benchmarkme 1.0.8 2022-06-12 [2] CRAN (R 4.5.0)
#> benchmarkmeData 1.0.4 2020-04-23 [2] CRAN (R 4.5.0)
#> bibtex 0.5.1 2023-01-26 [2] CRAN (R 4.5.0)
#> Biobase * 2.68.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocFileCache 2.16.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocGenerics * 0.54.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocIO 1.18.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocManager 1.30.25 2024-08-28 [2] CRAN (R 4.5.0)
#> BiocNeighbors 2.2.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocParallel 1.42.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocSingular 1.24.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocStyle * 2.36.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BiocVersion 3.21.1 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> Biostrings 2.76.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> BisqueRNA * 1.0.5 2021-05-23 [2] CRAN (R 4.5.0)
#> bit 4.6.0 2025-03-06 [2] CRAN (R 4.5.0)
#> bit64 4.6.0-1 2025-01-16 [2] CRAN (R 4.5.0)
#> bitops 1.0-9 2024-10-03 [2] CRAN (R 4.5.0)
#> blob 1.2.4 2023-03-17 [2] CRAN (R 4.5.0)
#> bluster 1.18.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> bookdown 0.43 2025-04-15 [2] CRAN (R 4.5.0)
#> bslib 0.9.0 2025-01-30 [2] CRAN (R 4.5.0)
#> cachem 1.1.0 2024-05-16 [2] CRAN (R 4.5.0)
#> circlize 0.4.16 2024-02-20 [2] CRAN (R 4.5.0)
#> cli 3.6.4 2025-02-13 [2] CRAN (R 4.5.0)
#> clue 0.3-66 2024-11-13 [2] CRAN (R 4.5.0)
#> cluster 2.1.8.1 2025-03-12 [3] CRAN (R 4.5.0)
#> codetools 0.2-20 2024-03-31 [3] CRAN (R 4.5.0)
#> colorspace 2.1-1 2024-07-26 [2] CRAN (R 4.5.0)
#> ComplexHeatmap 2.24.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> config 0.3.2 2023-08-30 [2] CRAN (R 4.5.0)
#> cowplot 1.1.3 2024-01-22 [2] CRAN (R 4.5.0)
#> crayon 1.5.3 2024-06-20 [2] CRAN (R 4.5.0)
#> curl 6.2.2 2025-03-24 [2] CRAN (R 4.5.0)
#> data.table 1.17.0 2025-02-22 [2] CRAN (R 4.5.0)
#> DBI 1.2.3 2024-06-02 [2] CRAN (R 4.5.0)
#> dbplyr 2.5.0 2024-03-19 [2] CRAN (R 4.5.0)
#> DeconvoBuddies * 1.0.0 2025-04-15 [1] Bioconductor 3.21 (R 4.5.0)
#> DelayedArray 0.34.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> digest 0.6.37 2024-08-19 [2] CRAN (R 4.5.0)
#> doParallel 1.0.17 2022-02-07 [2] CRAN (R 4.5.0)
#> dplyr * 1.1.4 2023-11-17 [2] CRAN (R 4.5.0)
#> dqrng 0.4.1 2024-05-28 [2] CRAN (R 4.5.0)
#> DT 0.33 2024-04-04 [2] CRAN (R 4.5.0)
#> edgeR 4.6.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> evaluate 1.0.3 2025-01-10 [2] CRAN (R 4.5.0)
#> ExperimentHub 2.16.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> farver 2.1.2 2024-05-13 [2] CRAN (R 4.5.0)
#> fastmap 1.2.0 2024-05-15 [2] CRAN (R 4.5.0)
#> filelock 1.0.3 2023-12-11 [2] CRAN (R 4.5.0)
#> foreach 1.5.2 2022-02-02 [2] CRAN (R 4.5.0)
#> generics * 0.1.3 2022-07-05 [2] CRAN (R 4.5.0)
#> GenomeInfoDb * 1.44.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> GenomeInfoDbData 1.2.14 2025-04-10 [2] Bioconductor
#> GenomicAlignments 1.44.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> GenomicRanges * 1.60.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> GetoptLong 1.0.5 2020-12-15 [2] CRAN (R 4.5.0)
#> ggbeeswarm 0.7.2 2023-04-29 [2] CRAN (R 4.5.0)
#> ggplot2 * 3.5.2 2025-04-09 [2] CRAN (R 4.5.0)
#> ggrepel 0.9.6 2024-09-07 [2] CRAN (R 4.5.0)
#> GlobalOptions 0.1.2 2020-06-10 [2] CRAN (R 4.5.0)
#> glue 1.8.0 2024-09-30 [2] CRAN (R 4.5.0)
#> golem 0.5.1 2024-08-27 [2] CRAN (R 4.5.0)
#> gridExtra 2.3 2017-09-09 [2] CRAN (R 4.5.0)
#> gtable 0.3.6 2024-10-25 [2] CRAN (R 4.5.0)
#> h5mread 1.0.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> HDF5Array 1.36.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> htmltools 0.5.8.1 2024-04-04 [2] CRAN (R 4.5.0)
#> htmlwidgets 1.6.4 2023-12-06 [2] CRAN (R 4.5.0)
#> httpuv 1.6.15 2024-03-26 [2] CRAN (R 4.5.0)
#> httr 1.4.7 2023-08-15 [2] CRAN (R 4.5.0)
#> igraph 2.1.4 2025-01-23 [2] CRAN (R 4.5.0)
#> IRanges * 2.42.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> irlba 2.3.5.1 2022-10-03 [2] CRAN (R 4.5.0)
#> iterators 1.0.14 2022-02-05 [2] CRAN (R 4.5.0)
#> jquerylib 0.1.4 2021-04-26 [2] CRAN (R 4.5.0)
#> jsonlite 2.0.0 2025-03-27 [2] CRAN (R 4.5.0)
#> KEGGREST 1.48.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> knitr 1.50 2025-03-16 [2] CRAN (R 4.5.0)
#> labeling 0.4.3 2023-08-29 [2] CRAN (R 4.5.0)
#> later 1.4.2 2025-04-08 [2] CRAN (R 4.5.0)
#> lattice 0.22-7 2025-04-02 [3] CRAN (R 4.5.0)
#> lazyeval 0.2.2 2019-03-15 [2] CRAN (R 4.5.0)
#> lifecycle 1.0.4 2023-11-07 [2] CRAN (R 4.5.0)
#> limma 3.64.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> limSolve 1.5.7.1 2024-02-05 [2] CRAN (R 4.5.0)
#> locfit 1.5-9.12 2025-03-05 [2] CRAN (R 4.5.0)
#> lpSolve 5.6.23 2024-12-14 [2] CRAN (R 4.5.0)
#> lubridate 1.9.4 2024-12-08 [2] CRAN (R 4.5.0)
#> magick 2.8.6 2025-03-23 [2] CRAN (R 4.5.0)
#> magrittr 2.0.3 2022-03-30 [2] CRAN (R 4.5.0)
#> MASS 7.3-65 2025-02-28 [3] CRAN (R 4.5.0)
#> Matrix 1.7-3 2025-03-11 [3] CRAN (R 4.5.0)
#> MatrixGenerics * 1.20.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> matrixStats * 1.5.0 2025-01-07 [2] CRAN (R 4.5.0)
#> memoise 2.0.1 2021-11-26 [2] CRAN (R 4.5.0)
#> metapod 1.16.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> mime 0.13 2025-03-17 [2] CRAN (R 4.5.0)
#> munsell 0.5.1 2024-04-01 [2] CRAN (R 4.5.0)
#> paletteer 1.6.0 2024-01-21 [2] CRAN (R 4.5.0)
#> pillar 1.10.2 2025-04-05 [2] CRAN (R 4.5.0)
#> pkgconfig 2.0.3 2019-09-22 [2] CRAN (R 4.5.0)
#> plotly 4.10.4 2024-01-13 [2] CRAN (R 4.5.0)
#> plyr 1.8.9 2023-10-02 [2] CRAN (R 4.5.0)
#> png 0.1-8 2022-11-29 [2] CRAN (R 4.5.0)
#> promises 1.3.2 2024-11-28 [2] CRAN (R 4.5.0)
#> purrr 1.0.4 2025-02-05 [2] CRAN (R 4.5.0)
#> quadprog 1.5-8 2019-11-20 [2] CRAN (R 4.5.0)
#> R6 2.6.1 2025-02-15 [2] CRAN (R 4.5.0)
#> rafalib 1.0.4 2025-04-08 [2] CRAN (R 4.5.0)
#> rappdirs 0.3.3 2021-01-31 [2] CRAN (R 4.5.0)
#> RColorBrewer 1.1-3 2022-04-03 [2] CRAN (R 4.5.0)
#> Rcpp 1.0.14 2025-01-12 [2] CRAN (R 4.5.0)
#> RCurl 1.98-1.17 2025-03-22 [2] CRAN (R 4.5.0)
#> RefManageR * 1.4.0 2022-09-30 [2] CRAN (R 4.5.0)
#> rematch2 2.1.2 2020-05-01 [2] CRAN (R 4.5.0)
#> reshape2 1.4.4 2020-04-09 [2] CRAN (R 4.5.0)
#> restfulr 0.0.15 2022-06-16 [2] CRAN (R 4.5.0)
#> rhdf5 2.52.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> rhdf5filters 1.20.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> Rhdf5lib 1.30.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> rjson 0.2.23 2024-09-16 [2] CRAN (R 4.5.0)
#> rlang 1.1.6 2025-04-11 [2] CRAN (R 4.5.0)
#> rmarkdown 2.29 2024-11-04 [2] CRAN (R 4.5.0)
#> Rsamtools 2.24.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> RSQLite 2.3.9 2024-12-03 [2] CRAN (R 4.5.0)
#> rsvd 1.0.5 2021-04-16 [2] CRAN (R 4.5.0)
#> rtracklayer 1.68.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> S4Arrays 1.8.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> S4Vectors * 0.46.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> sass 0.4.10 2025-04-11 [2] CRAN (R 4.5.0)
#> ScaledMatrix 1.16.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> scales 1.3.0 2023-11-28 [2] CRAN (R 4.5.0)
#> scater 1.36.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> scran 1.36.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> scuttle 1.18.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> sessioninfo * 1.2.3 2025-02-05 [2] CRAN (R 4.5.0)
#> shape 1.4.6.1 2024-02-23 [2] CRAN (R 4.5.0)
#> shiny 1.10.0 2024-12-14 [2] CRAN (R 4.5.0)
#> shinyWidgets 0.9.0 2025-02-21 [2] CRAN (R 4.5.0)
#> SingleCellExperiment * 1.30.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> SparseArray 1.8.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> SpatialExperiment * 1.18.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> spatialLIBD * 1.19.12 2025-04-08 [2] Bioconductor 3.21 (R 4.5.0)
#> statmod 1.5.0 2023-01-06 [2] CRAN (R 4.5.0)
#> stringi 1.8.7 2025-03-27 [2] CRAN (R 4.5.0)
#> stringr 1.5.1 2023-11-14 [2] CRAN (R 4.5.0)
#> SummarizedExperiment * 1.38.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> tibble * 3.2.1 2023-03-20 [2] CRAN (R 4.5.0)
#> tidyr * 1.3.1 2024-01-24 [2] CRAN (R 4.5.0)
#> tidyselect 1.2.1 2024-03-11 [2] CRAN (R 4.5.0)
#> timechange 0.3.0 2024-01-18 [2] CRAN (R 4.5.0)
#> UCSC.utils 1.4.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> utf8 1.2.4 2023-10-22 [2] CRAN (R 4.5.0)
#> vctrs 0.6.5 2023-12-01 [2] CRAN (R 4.5.0)
#> vipor 0.4.7 2023-12-18 [2] CRAN (R 4.5.0)
#> viridis 0.6.5 2024-01-29 [2] CRAN (R 4.5.0)
#> viridisLite 0.4.2 2023-05-02 [2] CRAN (R 4.5.0)
#> withr 3.0.2 2024-10-28 [2] CRAN (R 4.5.0)
#> xfun 0.52 2025-04-02 [2] CRAN (R 4.5.0)
#> XML 3.99-0.18 2025-01-01 [2] CRAN (R 4.5.0)
#> xml2 1.3.8 2025-03-14 [2] CRAN (R 4.5.0)
#> xtable 1.8-4 2019-04-21 [2] CRAN (R 4.5.0)
#> XVector 0.48.0 2025-04-15 [2] Bioconductor 3.21 (R 4.5.0)
#> yaml 2.3.10 2024-07-26 [2] CRAN (R 4.5.0)
#>
#> [1] /tmp/Rtmpw2uybS/Rinst16eac642cb921
#> [2] /home/biocbuild/bbs-3.21-bioc/R/site-library
#> [3] /home/biocbuild/bbs-3.21-bioc/R/library
#> * ── Packages attached to the search path.
#>
#> ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────12 Bibliography
This vignette was generated using BiocStyle (Oleś, 2025) with knitr (Xie, 2025) and rmarkdown (Allaire, Xie, Dervieux et al., 2024) running behind the scenes.
Citations made with RefManageR (McLean, 2017).
[1] J. Allaire, Y. Xie, C. Dervieux, et al. rmarkdown: Dynamic Documents for R. R package version 2.29. 2024. URL: https://github.com/rstudio/rmarkdown.
[2] L. A. Huuki-Myers, K. R. Maynard, S. C. Hicks, et al. DeconvoBuddies: a R/Bioconductor package with deconvolution helper functions. https://github.com/LieberInstitute/DeconvoBuddies/DeconvoBuddies - R package version 1.0.0. 2025. DOI: 10.18129/B9.bioc.DeconvoBuddies. URL: http://www.bioconductor.org/packages/DeconvoBuddies.
[3] M. W. McLean. “RefManageR: Import and Manage BibTeX and BibLaTeX References in R”. In: The Journal of Open Source Software (2017). DOI: 10.21105/joss.00338.
[4] A. Oleś. BiocStyle: Standard styles for vignettes and other Bioconductor documents. R package version 2.36.0. 2025. DOI: 10.18129/B9.bioc.BiocStyle. URL: https://bioconductor.org/packages/BiocStyle.
[5] R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria, 2025. URL: https://www.R-project.org/.
[6] H. Wickham. “testthat: Get Started with Testing”. In: The R Journal 3 (2011), pp. 5–10. URL: https://journal.r-project.org/archive/2011-1/RJournal_2011-1_Wickham.pdf.
[7] H. Wickham, W. Chang, R. Flight, et al. sessioninfo: R Session Information. R package version 1.2.3. 2025. DOI: 10.32614/CRAN.package.sessioninfo. URL: https://CRAN.R-project.org/package=sessioninfo.
[8] Y. Xie. knitr: A General-Purpose Package for Dynamic Report Generation in R. R package version 1.50. 2025. URL: https://yihui.org/knitr/.