Package 'quantiseqr' reference manual

Title:	Quantification of the Tumor Immune contexture from RNA-seq data
Description:	This package provides a streamlined workflow for the quanTIseq method, developed to perform the quantification of the Tumor Immune contexture from RNA-seq data. The quantification is performed against the TIL10 signature (dissecting the contributions of ten immune cell types), carefully crafted from a collection of human RNA-seq samples. The TIL10 signature has been extensively validated using simulated, flow cytometry, and immunohistochemistry data.
Authors:	Federico Marini [aut, cre] , Francesca Finotello [aut]
Maintainer:	Federico Marini <[email protected]>
License:	GPL-3
Version:	1.15.0
Built:	2024-10-31 03:42:45 UTC
Source:	https://github.com/bioc/quantiseqr

Check the signature matrix

Description

Checks requirements for the signature matrix, with respect to the expression matrix data provided (the one on which the deconvolution algorithm needs to be run)

Usage

check_signature(signature_matrix, mix_mat)
check_signature(signature_matrix, mix_mat)

Arguments

`signature_matrix`	A data.frame or a matrix object, containing the signature matrix
`mix_mat`	Mixture matrix, storing the information provided as `expression_data` to the main function, `run_quantiseqr()`

Details

Performs a number of checks to ensure the compatibility of the provided signature matrix in quantiseqr, referring also to the content of the mix_mat mixture matrix, to be deconvoluted

Value

Invisible NULL

An exemplary dataset with samples from four patients with metastatic melanoma

Description

An exemplary dataset with samples from four patients with metastatic melanoma

Details

quantiseqr ships with an example dataset with samples from four patients with metastatic melanoma. The dataset quantiseqr::dataset_racle contains

a gene expression matrix (dataset_racle$expr_mat) generated using bulk RNA-seq
'gold standard' estimates of immune cell contents profiled with FACS (dataset_racle$ref).

References

Racle et al, 2017 - https://doi.org/10.7554/eLife.26476.049

Solve Least Squares with Equality and Inequality Constraints (LSEI) problem

Description

Solve Least Squares with Equality and Inequality Constraints (LSEI) problem

Usage

DClsei(b, A, G, H, scaling = NULL)
DClsei(b, A, G, H, scaling = NULL)

Arguments

`b`	Numeric vector containing the right-hand side of the quadratic function to be minimized.
`A`	Numeric matrix containing the coefficients of the quadratic function to be minimized.
`G`	Numeric matrix containing the coefficients of the inequality constraints.
`H`	Numeric vector containing the right-hand side of the inequality constraints.
`scaling`	A vector of scaling factors to by applied to the estimates. Its length should equal the number of columns of A.

Details

The limSolve::lsei() function is used as underlying framework. Please refer to that function for more details.

Value

A vector containing the solution of the LSEI problem.

Examples

data(dataset_racle)
mixture <- dataset_racle$expr_mat
signature.file <- system.file(
  "extdata", "TIL10_signature.txt", package = "quantiseqr", mustWork = TRUE)
signature <- read.table(signature.file, header = TRUE, sep = "\t", row.names = 1)
scaling.file <- system.file(
  "extdata", "TIL10_mRNA_scaling.txt", package = "quantiseqr", mustWork = TRUE)
scaling <- as.vector(
  as.matrix(read.table(scaling.file, header = FALSE, sep = "\t", row.names = 1)))

cgenes <- intersect(rownames(signature), rownames(mixture))
b <- as.vector(as.matrix(mixture[cgenes,1, drop=FALSE]))
A <- as.matrix(signature[cgenes,])

G <- matrix(0, ncol = ncol(A), nrow = ncol(A))
diag(G) <- 1
G <- rbind(G, rep(-1, ncol(G)))
H <- c(rep(0, ncol(A)), -1)
# cellfrac <- quantiseqr:::DClsei(b = b, A = A, G= G, H = H, scaling = scaling)
data(dataset_racle)
mixture <- dataset_racle$expr_mat
signature.file <- system.file(
  "extdata", "TIL10_signature.txt", package = "quantiseqr", mustWork = TRUE)
signature <- read.table(signature.file, header = TRUE, sep = "\t", row.names = 1)
scaling.file <- system.file(
  "extdata", "TIL10_mRNA_scaling.txt", package = "quantiseqr", mustWork = TRUE)
scaling <- as.vector(
  as.matrix(read.table(scaling.file, header = FALSE, sep = "\t", row.names = 1)))

cgenes <- intersect(rownames(signature), rownames(mixture))
b <- as.vector(as.matrix(mixture[cgenes,1, drop=FALSE]))
A <- as.matrix(signature[cgenes,])

G <- matrix(0, ncol = ncol(A), nrow = ncol(A))
diag(G) <- 1
G <- rbind(G, rep(-1, ncol(G)))
H <- c(rep(0, ncol(A)), -1)
# cellfrac <- quantiseqr:::DClsei(b = b, A = A, G= G, H = H, scaling = scaling)

Perform robust regression

Description

Perform robust regression

Usage

DCrr(b, A, method = c("hampel", "huber", "bisquare"), scaling = NULL)
DCrr(b, A, method = c("hampel", "huber", "bisquare"), scaling = NULL)

Arguments

`b`	Numeric vector containing the right-hand side of the quadratic function to be minimized.
`A`	Numeric matrix containing the coefficients of the quadratic function to be minimized.
`method`	Character specifying the robust regression method to be used among deconvolution methods: "hampel", "huber", or "bisquare". Default: "hampel".
`scaling`	A vector of scaling factors to by applied to the estimates. Its length should equal the number of columns of A.

Details

The MASS::rlm() function is used as underlying framework. Please refer to that function for more details.

Value

A vector containing robust least-square estimates.

Examples

data(dataset_racle)
mixture <- dataset_racle$expr_mat
signature.file <- system.file(
  "extdata", "TIL10_signature.txt", package = "quantiseqr", mustWork = TRUE)
signature <- read.table(signature.file, header = TRUE, sep = "\t", row.names = 1)
scaling.file <- system.file(
  "extdata", "TIL10_mRNA_scaling.txt", package = "quantiseqr", mustWork = TRUE)
scaling <- as.vector(
  as.matrix(read.table(scaling.file, header = FALSE, sep = "\t", row.names = 1)))

cgenes <- intersect(rownames(signature), rownames(mixture))
b <- as.vector(as.matrix(mixture[cgenes,1, drop=FALSE]))
A <- as.matrix(signature[cgenes,])

# cellfrac <- quantiseqr:::DCrr(b = b, A = A, scaling = scaling)
data(dataset_racle)
mixture <- dataset_racle$expr_mat
signature.file <- system.file(
  "extdata", "TIL10_signature.txt", package = "quantiseqr", mustWork = TRUE)
signature <- read.table(signature.file, header = TRUE, sep = "\t", row.names = 1)
scaling.file <- system.file(
  "extdata", "TIL10_mRNA_scaling.txt", package = "quantiseqr", mustWork = TRUE)
scaling <- as.vector(
  as.matrix(read.table(scaling.file, header = FALSE, sep = "\t", row.names = 1)))

cgenes <- intersect(rownames(signature), rownames(mixture))
b <- as.vector(as.matrix(mixture[cgenes,1, drop=FALSE]))
A <- as.matrix(signature[cgenes,])

# cellfrac <- quantiseqr:::DCrr(b = b, A = A, scaling = scaling)

Convert a `Biobase::ExpressionSet` to a gene-expression matrix.

Description

Convert a Biobase::ExpressionSet to a gene-expression matrix.

Usage

eset_to_matrix(eset, column)
eset_to_matrix(eset, column)

Arguments

`eset`	`ExpressionSet`
`column`	column name of the `fData()` table, which contains the HGNC gene symbols.

Value

A matrix with gene symbols as rownames and sample identifiers as colnames.

Examples


data(dataset_racle)
dim(dataset_racle$expr_mat)

library("Biobase")
es_racle <- ExpressionSet(assayData = dataset_racle$expr_mat)
featureData(es_racle)$gene_symbol <- rownames(dataset_racle$expr_mat)

es_racle

head(eset_to_matrix(es_racle, "gene_symbol"))

data(dataset_racle)
dim(dataset_racle$expr_mat)

library("Biobase")
es_racle <- ExpressionSet(assayData = dataset_racle$expr_mat)
featureData(es_racle)$gene_symbol <- rownames(dataset_racle$expr_mat)

es_racle

head(eset_to_matrix(es_racle, "gene_symbol"))

Extract tumor immune quantifications

Description

Extract tumor immune quantifications from a SummarizedExperiment object, previously processed with run_quantiseqr()

Usage

extract_ti_from_se(se)
extract_ti_from_se(se)

Arguments

`se`	A `SummarizedExperiment` object, or any of its derivates, which contains the quantifications extracted via `quantiseqr` in its `colData` slot.

Value

A data.frame, formatted as required by downstream functions

Examples

data(dataset_racle)
dim(dataset_racle$expr_mat)

# using a SummarizedExperiment object
library("SummarizedExperiment")
se_racle <- SummarizedExperiment(
  assays = List(
    abundance = dataset_racle$expr_mat
  ),
  colData = DataFrame(
    SampleName = colnames(dataset_racle$expr_mat)
  )
)

res_run_SE <- quantiseqr::run_quantiseq(
    expression_data = se_racle,
    signature_matrix = "TIL10",
    is_arraydata = FALSE,
    is_tumordata = TRUE,
    scale_mRNA = TRUE
)

extract_ti_from_se(res_run_SE)

data(dataset_racle)
dim(dataset_racle$expr_mat)

# using a SummarizedExperiment object
library("SummarizedExperiment")
se_racle <- SummarizedExperiment(
  assays = List(
    abundance = dataset_racle$expr_mat
  ),
  colData = DataFrame(
    SampleName = colnames(dataset_racle$expr_mat)
  )
)

res_run_SE <- quantiseqr::run_quantiseq(
    expression_data = se_racle,
    signature_matrix = "TIL10",
    is_arraydata = FALSE,
    is_tumordata = TRUE,
    scale_mRNA = TRUE
)

extract_ti_from_se(res_run_SE)

Format the mixture matrix before deconvolution

Description

Format the mixture matrix before deconvolution

Usage

fixMixture(mix.mat, arrays = FALSE)
fixMixture(mix.mat, arrays = FALSE)

Arguments

`mix.mat`	Matrix or data.frame with RNA-seq gene TPM or microarray expression values for all samples to be deconvoluted, with gene symbols as row names and sample IDs as column names. Expression levels should be on non-log scale.
`arrays`	Logical value. Should be set to TRUE if the expression data are from microarrays. For RNA-seq data, this has to be FALSE (default value).

Value

The input matrix transformed to the natural scale (if needed), with fixed gene names on the rows, and TPM (for RNA-seq) or quantile (for microarrays) normalized.

Examples


data(dataset_racle)
# mixture.fix <- quantiseqr:::fixMixture(dataset_racle$expr_mat)
data(dataset_racle)
# mixture.fix <- quantiseqr:::fixMixture(dataset_racle$expr_mat)

Scale deconvoluted cell fractions to cell densities

Description

Scale deconvoluted cell fractions to cell densities

Usage

get_densities(DCres, density_info)
get_densities(DCres, density_info)

Arguments

`DCres`	Data.frame of deconvoluted cell fractions computed with the `run_quantiseq()` function, with sample identifiers as row names.
`density_info`	Named numeric vector of total cell densities per sample. The vector names should match the sample identifiers specified in DCres. These values are derived from the quantitative analysis of imaging data.

Value

A data.frame of cell densities, samples by cell types.

Examples

data(dataset_racle)
mixture <- dataset_racle$expr_mat

res_quantiseq_run <- quantiseqr::run_quantiseq(
  expression_data = dataset_racle$expr_mat,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = TRUE,
  scale_mRNA = TRUE
)

totcells <- rnorm(n = ncol(mixture), mean = 1e4)
names(totcells) <- colnames(mixture)
celldens <- get_densities(res_quantiseq_run, totcells)
data(dataset_racle)
mixture <- dataset_racle$expr_mat

res_quantiseq_run <- quantiseqr::run_quantiseq(
  expression_data = dataset_racle$expr_mat,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = TRUE,
  scale_mRNA = TRUE
)

totcells <- rnorm(n = ncol(mixture), mean = 1e4)
names(totcells) <- colnames(mixture)
celldens <- get_densities(res_quantiseq_run, totcells)

Perform quantile normalization of expression data

Description

Perform quantile normalization of expression data

Usage

makeQN(mix.mat)
makeQN(mix.mat)

Arguments

mix.mat

Matrix or data.frame with microarray gene expression values for all samples to be deconvoluted, with gene symbols as row names and sample IDs as column names. Expression levels should be on non-log scale.

Value

The input matrix transformed with quantile normalization.

Examples


data(dataset_racle)
# mixture.quantile <- quantiseqr:::makeQN(dataset_racle$expr_mat)
data(dataset_racle)
# mixture.quantile <- quantiseqr:::makeQN(dataset_racle$expr_mat)

Rename gene symbols before deconvolution

Description

Rename gene symbols before deconvolution

Usage

mapGenes(mydata)
mapGenes(mydata)

Arguments

mydata

Matrix or data.frame with RNA-seq gene TPM or microarray gene expression values for all samples to be deconvoluted, with gene symbols as row names and sample IDs as column names.

Value

The input matrix with updated gene names on the rows.

Examples


data(dataset_racle)
# mixture.fixgenes <- quantiseqr:::mapGenes(dataset_racle$expr_mat)
data(dataset_racle)
# mixture.fixgenes <- quantiseqr:::mapGenes(dataset_racle$expr_mat)

Plot the information on the tumor immune contexture

Description

Plot the information on the tumor immune contexture, as extracted with run_quantiseqr()

Usage

quantiplot(obj)
quantiplot(obj)

Arguments

obj

An object, either

a SummarizedExperiment where the quantifications are stored
a simple data.frame object, as obtained by run_quantiseqr()

Value

A ggplot object

Examples

data(dataset_racle)
dim(dataset_racle$expr_mat)
res_quantiseq_run <- quantiseqr::run_quantiseq(
  expression_data = dataset_racle$expr_mat,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = TRUE,
  scale_mRNA = TRUE
)

# using a SummarizedExperiment object
library("SummarizedExperiment")
se_racle <- SummarizedExperiment(
  assays = List(
    abundance = dataset_racle$expr_mat
  ),
  colData = DataFrame(
    SampleName = colnames(dataset_racle$expr_mat)
  )
)

res_run_SE <- quantiseqr::run_quantiseq(
    expression_data = se_racle,
    signature_matrix = "TIL10",
    is_arraydata = FALSE,
    is_tumordata = TRUE,
    scale_mRNA = TRUE
)

quantiplot(res_quantiseq_run)
# equivalent to...
quantiplot(res_run_SE)

data(dataset_racle)
dim(dataset_racle$expr_mat)
res_quantiseq_run <- quantiseqr::run_quantiseq(
  expression_data = dataset_racle$expr_mat,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = TRUE,
  scale_mRNA = TRUE
)

# using a SummarizedExperiment object
library("SummarizedExperiment")
se_racle <- SummarizedExperiment(
  assays = List(
    abundance = dataset_racle$expr_mat
  ),
  colData = DataFrame(
    SampleName = colnames(dataset_racle$expr_mat)
  )
)

res_run_SE <- quantiseqr::run_quantiseq(
    expression_data = se_racle,
    signature_matrix = "TIL10",
    is_arraydata = FALSE,
    is_tumordata = TRUE,
    scale_mRNA = TRUE
)

quantiplot(res_quantiseq_run)
# equivalent to...
quantiplot(res_run_SE)

Run quanTIseq deconvolution

Description

Run quanTIseq deconvolution

Usage

quanTIseq(currsig, currmix, scaling = TRUE, method = "lsei")
quanTIseq(currsig, currmix, scaling = TRUE, method = "lsei")

Arguments

`currsig`	Signature matrix to be used for deconvolution (format: genes by cell types).
`currmix`	Mixture matrix to be deconvoluted (format: genes by samples).
`scaling`	Logical value. If set to FALSE, it disables the correction of cell-type-specific mRNA content bias. Default: TRUE
`method`	Character string, defining the deconvolution method to be used: `lsei` for constrained least squares regression, `hampel`, `huber`, or `bisquare` for robust regression with Huber, Hampel, or Tukey bisquare estimators, respectively. Default: `lsei`.

Value

A data.frame of cell fractions, cell types by samples.

Examples


data(dataset_racle)
mixture <- dataset_racle$expr_mat
signature.file <- system.file(
  "extdata", "TIL10_signature.txt", package = "quantiseqr", mustWork = TRUE)
signature <- read.table(signature.file, header = TRUE, sep = "\t", row.names = 1)
# cellfrac <- quantiseqr:::quanTIseq(mixture, signature)
data(dataset_racle)
mixture <- dataset_racle$expr_mat
signature.file <- system.file(
  "extdata", "TIL10_signature.txt", package = "quantiseqr", mustWork = TRUE)
signature <- read.table(signature.file, header = TRUE, sep = "\t", row.names = 1)
# cellfrac <- quantiseqr:::quanTIseq(mixture, signature)

Helper functions for quanTIseq

Description

Helper functions for quanTIseq

quantiseqr package

Description

pkg description

Run the quanTIseq algorithm

Description

Use quanTIseq to deconvolute a gene expression matrix.

Usage

run_quantiseq(
  expression_data,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = FALSE,
  scale_mRNA = TRUE,
  method = "lsei",
  column = "gene_symbol",
  rm_genes = NULL,
  return_se = is(expression_data, "SummarizedExperiment")
)
run_quantiseq(
  expression_data,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = FALSE,
  scale_mRNA = TRUE,
  method = "lsei",
  column = "gene_symbol",
  rm_genes = NULL,
  return_se = is(expression_data, "SummarizedExperiment")
)

Arguments

`expression_data`	The gene expression information, containing the TPM values for the measured features. Can be provided as a simple gene expression matrix, or a data frame (with HGNC gene symbols as row names and sample identifiers as column names) an `ExpressionSet` object (from the Biobase package), where the HGNC gene symbols are provided in a column of the `fData` slot - that is specified by the `column` parameter below a `SummarizedExperiment` object, or any of the derivative classes (e.g. DESeq2's `DESeqDataSet`), in which the assay (default: "abundance") is containing the TPMs as expected. Internally, `quantiseqr` handles the conversion to an object which is used in the deconvolution procedure.
`signature_matrix`	Character string, specifying the name of the signature matrix. At the moment, only the original `TIL10` signature can be selected.
`is_arraydata`	Logical value. Should be set to TRUE if the expression data are originating from microarray data. For RNA-seq data, this has to be FALSE (default value). If set to TRUE, the `rmgenes` parameter (see below) is set to "none".
`is_tumordata`	Logical value. Should be set to TRUE if the expression data is from tumor samples. Default: FALSE (e.g. for RNA-seq from blood samples)
`scale_mRNA`	Logical value. If set to FALSE, it disables the correction of cell-type-specific mRNA content bias. Default: TRUE
`method`	Character string, defining the deconvolution method to be used: `lsei` for constrained least squares regression, `hampel`, `huber`, or `bisquare` for robust regression with Huber, Hampel, or Tukey bisquare estimators, respectively. Default: `lsei`.
`column`	Character, specifies which column in the `fData` slot (for the ExpressionSet object) contains the information of the HGNC gene symbol identifiers
`rm_genes`	Character vector, specifying which genes have to be excluded from the deconvolution analysis. It can be provided as a vector of gene symbols (contained in the `expression_data`) a single string among the choices of "none" (no genes are removed) and "default" (a list of genes with noisy expression RNA-seq data is removed, as explained in the quanTIseq paper). Default: "default" for RNA-seq data, "none" for microarrays.
`return_se`	Logical value, controls the format of how the quantification is returned. If providing a `SummarizedExperiment`, it can simply extend its `colData` component, without the need to create a separate data frame as output.

Details

The values contained in the expression_data need to be provided as TPM values, as this is the format also used to store the TIL10 signature, upon which quanTIseq builds to perform the immune cell type deconvolution. Expression data should not be provided in logarithmic scale.

If providing the expression_data as a SummarizedExperiment/DESeqDataSet object, it might be beneficial that this has been created via tximport - if this is the case, the assay named "abundance" will be automatically created upon importing the transcript quantification results.

Value

A data.frame containing the quantifications of the cell type proportions, or alternatively, if providing expression_data as SummarizedExperiment and setting return_se to TRUE, a SummarizedExperiment with the quantifications included by expanding the colData slot of the original object

References

F. Finotello, C. Mayer, C. Plattner, G. Laschober, D. Rieder, H. Hackl, A. Krogsdam, Z. Loncova, W. Posch, D. Wilflingseder, S. Sopper, M. Jsselsteijn, T. P. Brouwer, D. Johnsons, Y. Xu, Y. Wang, M. E. Sanders, M. V. Estrada, P. Ericsson-Gonzalez, P. Charoentong, J. Balko, N. F. d. C. C. de Miranda, Z. Trajanoski. "Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data". Genome Medicine 2019;11(1):34. doi: 10.1186/s13073-019-0638-6.

C. Plattner, F. Finotello, D. Rieder. "Chapter Ten - Deconvoluting tumor-infiltrating immune cells from RNA-seq data using quanTIseq". Methods in Enzymology, 2020. doi: 10.1016/bs.mie.2019.05.056.

Examples


data(dataset_racle)
dim(dataset_racle$expr_mat)
res_quantiseq_run <- quantiseqr::run_quantiseq(
  expression_data = dataset_racle$expr_mat,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = TRUE,
  scale_mRNA = TRUE
)

# using a SummarizedExperiment object
library("SummarizedExperiment")
se_racle <- SummarizedExperiment(
  assays = List(
    abundance = dataset_racle$expr_mat
  ),
  colData = DataFrame(
    SampleName = colnames(dataset_racle$expr_mat)
  )
)

res_run_SE <- quantiseqr::run_quantiseq(
    expression_data = se_racle,
    signature_matrix = "TIL10",
    is_arraydata = FALSE,
    is_tumordata = TRUE,
    scale_mRNA = TRUE
)

data(dataset_racle)
dim(dataset_racle$expr_mat)
res_quantiseq_run <- quantiseqr::run_quantiseq(
  expression_data = dataset_racle$expr_mat,
  signature_matrix = "TIL10",
  is_arraydata = FALSE,
  is_tumordata = TRUE,
  scale_mRNA = TRUE
)

# using a SummarizedExperiment object
library("SummarizedExperiment")
se_racle <- SummarizedExperiment(
  assays = List(
    abundance = dataset_racle$expr_mat
  ),
  colData = DataFrame(
    SampleName = colnames(dataset_racle$expr_mat)
  )
)

res_run_SE <- quantiseqr::run_quantiseq(
    expression_data = se_racle,
    signature_matrix = "TIL10",
    is_arraydata = FALSE,
    is_tumordata = TRUE,
    scale_mRNA = TRUE
)

SummarizedExperiment to matrix

Description

SummarizedExperiment to matrix

Usage

se_to_matrix(se, assay = "abundance")
se_to_matrix(se, assay = "abundance")

Arguments

`se`	A `SummarizedExperiment` object, or any of its derivates, which contains the information on the TPM expression values, which are stored in a specified assay slot.
`assay`	A character string, specifying the name of the `assays` component of the `se` object. Defaults to "abundance", as this is the common convention used e.g. by the `tximport` package to store the values imported from the transcript level quantifications

Value

A matrix object, containing the TPM values, ready to be used in the framework of quantiseqr

Examples

library("SummarizedExperiment")
library("macrophage")
data("gse", package = "macrophage")
se <- gse

# If using ENSEMBL or Gencode gene annotation, you might want to convert the row names
## in this case, the gene symbols are provided as rowData information
rownames(se) <- rowData(se)$SYMBOL

tpm_matrix <- se_to_matrix(se, assay = "abundance")

## otherwise, you can map the identifiers via
library("org.Hs.eg.db")
library("AnnotationDbi")
se <- gse
# keep the parts before the '.', used in the Gencode annotation
rownames(se) <- substr(rownames(se), 1, 15)
gene_names <- mapIds(org.Hs.eg.db,
                     keys = rownames(se),
                     column = "SYMBOL",
                     keytype = "ENSEMBL")
rownames(se) <- gene_names

# If you require to convert the counts to TPMs by hand, you need a vector of
# gene lengths as well, and then run this simple function on the count matrix
counts_to_tpm <- function(counts, lengths) {
  ratio <- counts / lengths
  mytpm <- ratio / sum(ratio) * 1e6
  return(mytpm)
}
# then run via
# tpmdata <- counts_to_tpm(count_matrix, genelength_vector)

library("SummarizedExperiment")
library("macrophage")
data("gse", package = "macrophage")
se <- gse

# If using ENSEMBL or Gencode gene annotation, you might want to convert the row names
## in this case, the gene symbols are provided as rowData information
rownames(se) <- rowData(se)$SYMBOL

tpm_matrix <- se_to_matrix(se, assay = "abundance")

## otherwise, you can map the identifiers via
library("org.Hs.eg.db")
library("AnnotationDbi")
se <- gse
# keep the parts before the '.', used in the Gencode annotation
rownames(se) <- substr(rownames(se), 1, 15)
gene_names <- mapIds(org.Hs.eg.db,
                     keys = rownames(se),
                     column = "SYMBOL",
                     keytype = "ENSEMBL")
rownames(se) <- gene_names

# If you require to convert the counts to TPMs by hand, you need a vector of
# gene lengths as well, and then run this simple function on the count matrix
counts_to_tpm <- function(counts, lengths) {
  ratio <- counts / lengths
  mytpm <- ratio / sum(ratio) * 1e6
  return(mytpm)
}
# then run via
# tpmdata <- counts_to_tpm(count_matrix, genelength_vector)

quanTIseq output for the simulation data of 1700 mixtures for RNA-seq data

Description

quanTIseq output for the simulation data of 1700 mixtures for RNA-seq data

Details

quanTIseq output for the simulation data of 1700 mixtures for RNA-seq data, stored as a data.frame with 1700 rows (all the single instances of the different mixtures) as returned by run_quantiseq(). Column names, accordingly, contain the names of the component of the TIL10 signature, namely "B.cells", "Macrophages.M1", "Macrophages.M2", "Monocytes", "Neutrophils", "NK.cells", "T.cells.CD4", "T.cells.CD8", "Tregs", "Dendritic.cells", and "Other" (indicating for example a proxy for the amount of tumor tissue).

This can be compared (see Vignette for an example) to the ground truth information on the components of the mixtures.

References

Finotello, F., Mayer, C., Plattner, C. et al. Correction to: Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data. Genome Med 11, 50 (2019). https://doi.org/10.1186/s13073-019-0655-5

Package 'quantiseqr'

Help Index

Check the signature matrix

Description

Usage

Arguments

Details

Value

An exemplary dataset with samples from four patients with metastatic melanoma

Description

Details

References

Solve Least Squares with Equality and Inequality Constraints (LSEI) problem

Description

Usage

Arguments

Details

Value

Examples

Perform robust regression

Description

Usage

Arguments

Details

Value

Examples

Convert a Biobase::ExpressionSet to a gene-expression matrix.

Description

Usage

Arguments

Value

Examples

Extract tumor immune quantifications

Description

Usage

Arguments

Value

Examples

Format the mixture matrix before deconvolution

Description

Usage

Arguments

Value

Examples

Scale deconvoluted cell fractions to cell densities

Description

Usage

Arguments

Value

Examples

Perform quantile normalization of expression data

Description

Usage

Arguments

Value

Examples

Rename gene symbols before deconvolution

Description

Usage

Arguments

Value

Examples

Plot the information on the tumor immune contexture

Description

Usage

Arguments

Value

Examples

Run quanTIseq deconvolution

Description

Usage

Arguments

Value

Examples

Helper functions for quanTIseq

Description

quantiseqr package

Description

Run the quanTIseq algorithm

Description

Convert a `Biobase::ExpressionSet` to a gene-expression matrix.