Package 'scQTLtools'

Title: scQTLtools: an R/Bioconductor package for comprehensive identification and visualization of single-cell eQTLs
Description: scQTLtools is a comprehensive R/Bioconductor package that facilitates end-to-end single-cell eQTL analysis, from preprocessing to visualization
Authors: Xiaofeng Wu [aut, cre, cph] (ORCID: <https://orcid.org/0009-0003-6254-5575>), Xin Huang [aut, cph] (ORCID: <https://orcid.org/0009-0005-2755-0357>), Jingtong Kang [com] (ORCID: <https://orcid.org/0009-0008-8343-3456>), Siwen Xu [aut, cph] (ORCID: <https://orcid.org/0000-0001-7936-0639>)
Maintainer: Xiaofeng Wu <[email protected]>
License: MIT + file LICENSE
Version: 1.5.0
Built: 2026-05-31 06:30:59 UTC
Source: https://github.com/bioc/scQTLtools

Help Index

Adjust p-values and filter SNP–gene pairs based on the adjusted p-values.

Description

Adjust p-values and filter SNP–gene pairs based on the adjusted p-values.

Usage

adjust_pvalues(result, pAdjustMethod = "bonferroni", pAdjustThreshold = 0.05)

Arguments

result

A data frame that contains information of SNP–gene pairs and raw p values.
pAdjustMethod

Method for p-value adjustment. One of "bonferroni", "holm", "hochberg", "hommel" or "BH". The default option is "bonferroni".
pAdjustThreshold

Only SNP–gene pairs with adjusted p-values meeting the threshold will be displayed. Default by 0.05.

Value

A data frame with adjusted p-values, filtered by threshold, containing information on SNP–gene pairs.

Examples

example_data <- data.frame(
  gene = c("Gene1", "Gene2", "Gene3", "Gene4"),
  SNP = c("SNP1", "SNP2", "SNP3", "SNP4"),
  pvalue = c(0.001, 0.04, 0.03, 0.0005))
pAdjustMethod <- "BH"
pAdjustThreshold <- 0.05
adjusted_result <- adjust_pvalues(example_data, pAdjustMethod,
pAdjustThreshold)

Build a ZINB model.

Description

Build a ZINB model.

Usage

buildZINB(counts)

Arguments

counts

A numeric vector of gene expression values.

Value

A list containing four parameters estimated from the ZINB model.

Examples

data(GeneData)
gene <- unlist(GeneData[1, ])
result <-buildZINB(gene)

Identify single-cell eQTLs.

Description

This function detects eQTLs using scRNA-seq data and genotype data.

Usage

callQTL(
  eQTLObject,
  gene_ids = NULL,
  downstream = NULL,
  upstream = NULL,
  gene_mart = NULL,
  snp_mart = NULL,
  pAdjustMethod = "bonferroni",
  useModel = "zinb",
  pAdjustThreshold = 0.05,
  logfcThreshold = 0.1
)

Arguments

eQTLObject

An S4 object of class eQTLObject.
gene_ids

A gene ID or a list of gene IDS.
downstream

Distance (in base pairs) downstream of the gene start site to search for associated SNPs.
upstream

Distance (in base pairs) upstream of the gene end site to search for associated SNPs.
gene_mart

A Mart object representing the BioMart gene database. If NULL, the Ensembl Gene BioMart will be used.
snp_mart

A Mart object representing the BioMart SNP database. If NULL, the Ensembl SNP BioMart will be used.
pAdjustMethod

Method used for multiple testing correction. One of "bonferroni", "holm", "hochberg", "hommel", or "BH". Default is "bonferroni".
useModel

Model used for fitting. One of "poisson", "zinb", or "linear."
pAdjustThreshold

Only SNP–gene pairs with adjusted p-values below the threshold will be retained. Default is 0.05.
logfcThreshold

The minimum beta coefficient required to report a SNP–gene pair as an eQTL.

Value

A data frame in which each row corresponds to a detected SNP–gene eQTL pair, including statistical and model fitting results.

Examples

data(EQTL_obj)
library(biomaRt)
gene_mart <- useEnsembl(biomart = "genes",
                        dataset = "hsapiens_gene_ensembl",
                        mirror = 'asia')
snp_mart <- useEnsembl(biomart = "snps",
                        dataset = "hsapiens_snp",
                        mirror = 'asia')
eqtl <- callQTL(
  eQTLObject = EQTL_obj,
  gene_ids = NULL,
  downstream = NULL,
  upstream = NULL,
  gene_mart = gene_mart,
  snp_mart = snp_mart,
  pAdjustMethod = 'bonferroni',
  useModel = 'linear',
  pAdjustThreshold = 0.05,
  logfcThreshold = 0.025
)

Validate SNP IDs in the input genotype matrix.

Description

Validate SNP IDs in the input genotype matrix.

Usage

checkSNPList(snpList, snp_mart = NULL, snpDataset = "hsapiens_snp")

Arguments

snpList

A list of SNPs IDs.
snp_mart

An object of class Mart representing the BioMart SNP database to connect to. If provided, this should be a Mart object obtained by calling useEnsembl(), which allows specifying a mirror in case of connection issues. If NULL, the function will create and use a Mart object pointing to the Ensembl SNP BioMart, using the specified snpDataset and a default mirror.
snpDataset

A character string specifying the SNP dataset to use from Ensembl. Default is hsapiens_snp for human SNPs.

Value

A data frame containing the genomic locations of the valid SNPs.

Examples

data(SNPData2)
snpList <- rownames(SNPData2)
snpDataset <- 'hsapiens_snp'
snps_loc <- checkSNPList(snpList = snpList,
                        snpDataset = snpDataset)

Normalize gene expression using CPM.

Description

CPM_normalize() applies Counts Per Million (CPM) normalization to a raw gene expression matrix.

Usage

CPM_normalize(expressionMatrix)

Arguments

expressionMatrix

A numeric matrix of raw gene expression counts, with genes as rows and cells as columns.

Value

A normalized gene expression matrix after applying CPM normalization.

Examples

data(GeneData)
CPM_normalize(GeneData)

Create a gene location data frame.

Description

Create a gene location data frame.

Usage

createGeneLoc(
  geneList,
  gene_mart = NULL,
  geneDataset = "hsapiens_gene_ensembl",
  OrgDb
)

Arguments

geneList

A gene ID or a list of genes IDs.
gene_mart

An object of class Mart representing the BioMart gene database to connect to. If provided, this should be a Mart object obtained by calling useEnsembl(), which allows specifying a mirror in case of connection issues. If NULL, the function will create and use a Mart object pointing to the Ensembl Gene BioMart, using the specified geneDataset and a default mirror.
geneDataset

A character string specifying the gene dataset to use from Ensembl. Default is hsapiens_gene_ensembl for human genes.
OrgDb

The name of the OrgDb package to use for gene annotation. Supported values include org.Hs.eg.db and org.Mm.eg.db.

Value

A data.frame containing gene location information.

Examples

data(GeneData)
geneList <- rownames(GeneData)
library(yulab.utils)
library(biomaRt)
OrgDb <- load_OrgDb("org.Hs.eg.db")
gene_mart <- useEnsembl(biomart = "genes",
                        dataset = "hsapiens_gene_ensembl",
                        mirror = 'asia')
gene_loc <- createGeneLoc(geneList = geneList,
                          gene_mart = gene_mart,
                          OrgDb = OrgDb)

Create an eQTLObject for storing sc-eQTL analysis data.

Description

This function creates an S4 object to store genotype and expression data, along with sample grouping and metadata for eQTL analysis.

Usage

createQTLObject(
  snpMatrix,
  genedata,
  biClassify = FALSE,
  species = NULL,
  group = NULL,
  ...
)

Arguments

snpMatrix

A genotype matrix where each row is a snp and each column is a cell. Encoding should be 0, 1, 2, 3.
genedata

A gene expression matrix, or a Seurat object or SingleCellExperiment object.
biClassify

Logical; whether to convert genotype encoding in snpMatrix to 0, 1, and 2. TRUE indicates conversion; FALSE indicates no conversion (default).
species

The species that the user wants to select, human or mouse.
group

A column name in the metadata of the Seurat or SingleCellExperiment object indicating cell groups (e.g., celltype or condition).
...

other parameters

Value

An S4 object of class eQTLObject.

Examples

data(SNPData)
data(GeneData)
eqtl <- createQTLObject(snpMatrix = SNPData,
                     genedata = GeneData,
                     biClassify = FALSE,
                     species = 'human',
                     group = NULL)

Create SNP location dataframe.

Description

Create SNP location dataframe.

Usage

createSNPsLoc(snpList, snp_mart = NULL, snpDataset = "hsapiens_snp")

Arguments

snpList

A list of SNPs IDs.
snp_mart

An object of class Mart representing the BioMart SNP database to connect to. If provided, this should be a Mart object obtained by calling useEnsembl(), which allows specifying a mirror in case of connection issues. If NULL, the function will create and use a Mart object pointing to the Ensembl SNP BioMart, using the specified snpDataset and a default mirror.
snpDataset

A character string specifying the SNP dataset to use from Ensembl. Default is hsapiens_snp for human SNPs.

Value

A data frame containing the SNP genomic locations.

Examples

snpList <- c('rs546', 'rs549', 'rs568', 'rs665', 'rs672')
library(biomaRt)
snp_mart <- useEnsembl(biomart = "snps",
                        dataset = "hsapiens_snp",
                        mirror = 'asia')
snp_loc <- createSNPsLoc(snpList = snpList,
                         snp_mart = snp_mart)

The package includes five examplt datasets: a gene expression matrix, two SNP genotype matrices, a Seurat object, and an eQTL object.

Description

GeneData: A dataset containing example gene expression data. Each row represents a gene and each column represents a cell.

SNPData: A dataset containing single nucleotide (SNP) data. Each row represents a variant and each column represents a cell.

Seurat_obj: A Seurat object containing GeneData along with associated metadata stored in the meta.data slot.

SNPData2: A smaller SNP dataset containing fewer variants and cells. Each row represents a variant and each column represents a cell.

EQTL_obj: An eQTLObject created using createQTLObject, where the raw expression matrix is normalized using normalizeGene, and both the genotype matrix and normalized expression matrix were filtered using filterGeneSNP.

Format

A numeric matrix with 100 rows (genes) and 2705 columns (cells).

A numeric matrix with 1000 rows (SNPs) and 2705 columns (cells).

An object of class Seurat.

A numeric matrix with 500 rows (SNPs) and 500 columns (cells).

See Also

createQTLObject, normalizeGene and filterGeneSNP, for the underlying functions that do the work.

An object of class eQTLObject.

Normalize the gene expression matrix with DESeq2.

Description

DESeq_normalize() normalizes a raw gene expression matrix using the DESeq2 package.

Usage

DESeq_normalize(expressionMatrix)

Arguments

expressionMatrix

A numeric matrix of raw gene expression counts, with genes as rows and cells as columns.

Value

A normalized gene expression matrix after applying DESeq normalization.

Examples

data(GeneData)
DESeq_normalize(GeneData)

Generate a boxplot of gene expression by SNP genotype.

Description

draw_boxplot() visualizes gene expression levels across different SNP genotypes using a boxplot.

Usage

draw_boxplot(df, unique_group)

Arguments

df

A data frame containing gene expression values, SNP genotypes, and group labels.
unique_group

A character string indicating the unique group name.

Value

ggplot

Examples

set.seed(123)
counts_Ref <- rnorm(50, mean = 10, sd = 2)
counts_Alt <- rnorm(50, mean = 12, sd = 2)
i <- rep("GroupA", 100)
unique_group <- unique(i)
dataframe <- data.frame(
expression = c(counts_Ref, counts_Alt),
snp = c(rep("REF", length(counts_Ref)),
        rep("ALT", length(counts_Alt))), group = i)
dataframe$snp <- factor(dataframe$snp, levels = c("REF", "ALT"))
draw_boxplot(df = dataframe, unique_group = unique_group)

Generate a histogram of gene expression by SNP genotype.

Description

draw_histplot() shows the distribution of expression values for each SNP genotype using histograms.

Usage

draw_histplot(df, unique_group)

Arguments

df

A data frame containing gene expression values, SNP genotypes, and group labels.
unique_group

A character string indicating the unique group name.

Value

ggplot

Examples

set.seed(123)
counts_Ref <- rnorm(50, mean = 10, sd = 2)
counts_Alt <- rnorm(50, mean = 12, sd = 2)
i <- rep("GroupA", 100);unique_group <- unique(i)
dataframe <- data.frame(expression = c(counts_Ref, counts_Alt),
                        snp = c(rep("REF", length(counts_Ref)),
                        rep("ALT", length(counts_Alt))), group = i)
dataframe$snp <- factor(dataframe$snp, levels = c("REF", "ALT"))
draw_histplot(df = dataframe, unique_group = unique_group)

Create a combined violin-box-scatter plot.

Description

draw_QTLplot() creates a composite plot that overlays violin plots, boxplots, and scatter points to illustrate the distribution and variability of gene expression across SNP groups.

Usage

draw_QTLplot(df, unique_group)

Arguments

df

A data frame containing gene expression values, SNP genotypes, and group labels.
unique_group

A character string indicating the unique group name.

Value

ggplot

Examples

set.seed(123)
counts_Ref <- rnorm(50, mean = 10, sd = 2)
counts_Alt <- rnorm(50, mean = 12, sd = 2)
i <- rep("GroupA", 100);unique_group <- unique(i)
dataframe <- data.frame(expression = c(counts_Ref, counts_Alt),
                        snp = c(rep("REF", length(counts_Ref)),
                        rep("ALT", length(counts_Alt))), group = i)
dataframe$snp <- factor(dataframe$snp, levels = c("REF", "ALT"))
draw_QTLplot(df = dataframe, unique_group = unique_group)

Generate a violin plot of gene expression by SNP genotype.

Description

draw_violinplot() visualizes gene expression levels across different SNP genotypes using violin plots.

Usage

draw_violinplot(df, unique_group)

Arguments

df

A data frame containing gene expression values, SNP genotypes, and group labels.
unique_group

A character string indicating the unique group name.

Value

ggplot

Examples

set.seed(123)
counts_Ref <- rnorm(50, mean = 10, sd = 2)
counts_Alt <- rnorm(50, mean = 12, sd = 2)
i <- rep("GroupA", 100);unique_group <- unique(i)
dataframe <- data.frame(expression = c(counts_Ref, counts_Alt),
                        snp = c(rep("REF", length(counts_Ref)),
                        rep("ALT", length(counts_Alt))), group = i)
dataframe$snp <- factor(dataframe$snp, levels = c("REF", "ALT"))
draw_violinplot(df = dataframe, unique_group = unique_group)

Class eQTLObject

Description

The eQTLObject class is designed to store data related to eQTL analysis, including data lists, result data frames, and additional metadata such as classification, species, and grouping information.

Value

An S4 object of class eQTLObject.

Slots

rawData

A gene expression dataframe, the row names represent gene IDs and the column names represent cell IDs.

filterData

Gene expression matrix after normalizing.

eQTLResult

The result dataframe obtained the sc-eQTL results.

biClassify

Logical; whether to convert genotype encoding in snpMatrix to 0, 1, and 2. TRUE indicates conversion; FALSE indicates no conversion (default).

species

The species that the user wants to select, human or mouse.

groupBy

Options for cell grouping, users can choose celltype, cellstatus, etc., depending on metadata.

useModel

model for fitting dataframe.

Filter a data frame of SNP–gene pairs by absolute beta.

Description

Filter a data frame of SNP–gene pairs by absolute beta.

Usage

filter_by_abs_b(result, logfcThreshold)

Arguments

result

A data frame that containing SNP–gene pairs and their beta values (b-value).
logfcThreshold

A numeric value specifying the minimum absolute b-value to retain. Default is 0.1.

Value

A data frame containing only rows with absolute b-values above the specified threshold.

Examples

example_result <- data.frame(
  gene = c("Gene1", "Gene2", "Gene3", "Gene4"),
  SNP = c("SNP1", "SNP2", "SNP3", "SNP4"),
  b = c(-2.5, 1.0, -0.5, 3.0))
logfcThreshold <- 0.1
filtered_result <- filter_by_abs_b(example_result, logfcThreshold)

Filter gene expression and genotype matrices by cell percentage thresholds.

Description

Filter gene expression and genotype matrices by cell percentage thresholds.

Usage

filterGeneSNP(
  eQTLObject,
  snpNumOfCellsPercent = 10,
  expressionMin = 0,
  expressionNumOfCellsPercent = 10
)

Arguments

eQTLObject

An S4 object of class eQTLObject.
snpNumOfCellsPercent

Numeric. Minimum percentage of cells required for each SNP genotype (e.g., AA, AG, and GG). Only SNPs where each genotype occurs in at least this proportion of cells are retained. Default is 10.
expressionMin

Numeric. Expression threshold used in combination with expressionNumOfCellsPercent to filter lowly expressed genes. Default is 0.
expressionNumOfCellsPercent

Numeric. Minimum percentage of cells in which a gene's expression must exceed expressionMin for the gene to be retained. Default is 10.

Value

An updated eQTLObject with filtered gene expression and SNP matrices.

Examples

data(SNPData)
data(GeneData)
eqtl <- createQTLObject(snpMatrix = SNPData, genedata = GeneData)
eqtl <- normalizeGene(eqtl)
eqtl <- filterGeneSNP(eqtl,
  snpNumOfCellsPercent = 2,
  expressionMin = 0,
  expressionNumOfCellsPercent = 2)

Retrieve Cells by SNP Value

Description

This function extracts the names of cells from a SNP matrix that correspond to a specified value for a given SNP.

Usage

get_cell_groups(snpMatrix, SNPid, biClassify)

Arguments

snpMatrix

A genotype matrix where each row is a snp and each column is a cell. Encoding should be 0, 1, 2, 3.
SNPid

A character string or numeric index representing the specific SNP of interest in the SNP matrix.
biClassify

Logical; whether to convert genotype encoding in snpMatrix to 0, 1, and 2. TRUE indicates conversion; FALSE indicates no conversion (default).

Value

A list of character vectors. Each vector contains the names of cells (i.e., column names of snpMatrix) corresponding to a specific genotype value at the given SNP.

Examples

data(SNPData)
biClassify <- FALSE
get_cell_groups(SNPData, "1:632445", biClassify)

Extract Counts from an Expression Matrix

Description

This function retrieves expression counts for a specified gene from an expression matrix, based on the provided list of cells.

Usage

get_counts(expressionMatrix, Geneid, cells)

Arguments

expressionMatrix

A numeric matrix of gene expression counts, with genes as rows and cells as columns.
Geneid

A character string or numeric index representing the specific gene of interest in expressionMatrix.
cells

A character vector of cell names (column names of expressionMatrix) from which to extract counts for the specified gene.

Value

A numeric vector containing the expression counts of the specified gene in the selected cells.

Examples

data(GeneData)
get_counts(GeneData, "CNN2",
          c("CGGCAGTGTAGCCCTG", "GGAGGATTCCCGTTCA"))

Access filtered data stored in an eQTLObject.

Description

Access filtered data stored in an eQTLObject.

Method to access eQTLObject filter data.

Usage

get_filter_data(x)

## S4 method for signature 'eQTLObject'
get_filter_data(x)

Arguments

x

An eQTLObject.

Value

Filtered matrices.

Filtered matrices.

Examples

data(EQTL_obj)
  get_filter_data(EQTL_obj)

Access model specification from an eQTLObject

Description

Access model specification from an eQTLObject

Method to access eQTLObject used model information.

Usage

get_model_info(x)

## S4 method for signature 'eQTLObject'
get_model_info(x)

Arguments

x

An eQTLObject.

Value

A character string indicating the model.

used model information of eQTLObject.

Examples

data(EQTL_obj)
  get_model_info(EQTL_obj)

Access raw data stored in an eQTLObject.

Description

Access raw data stored in an eQTLObject.

Method to access eQTLObject raw data.

Usage

get_raw_data(x)

## S4 method for signature 'eQTLObject'
get_raw_data(x)

Arguments

x

An eQTLObject.

Value

raw data matrix.

raw data matrix.

Examples

data(EQTL_obj)
  get_raw_data(EQTL_obj)

Access eQTLs results from an eQTLObject.

Description

Access eQTLs results from an eQTLObject.

Method to access the result of identifying eQTLs.

Usage

get_result_info(x)

## S4 method for signature 'eQTLObject'
get_result_info(x)

Arguments

x

An eQTLObject.

Value

A data frame where each row corresponds to an identified SNP–gene pair.

A data frame with eQTL results.

Examples

data(EQTL_obj)
  get_result_info(EQTL_obj)

Progress Bar for Model Analysis.

Description

Initializes a progress bar to track the computation progress during model fitting procedures such as linearModel, poissonModel, and zinbModel. This helps users monitor the status of long-running analyses.

Usage

initialize_progress_bar(total, k)

Arguments

total

Integer. The total number of iterations or steps to be completed in the analysis.
k

Character. An identifier or label for the specific group or subset being analyzed, used to annotate progress messages.

Value

An object of class progress_bar from the progress package, which can be updated using the $tick() method.

Examples

unique_group <- c("CMP", "GMP")
total_snp_count <- 10  # assume each group have 100 SNP.
pb_model <- lapply(unique_group, function(k) {
    pb <- initialize_progress_bar(total = total_snp_count, k)
    for (i in seq_len(total_snp_count)) {
        Sys.sleep(0.1)  # assume progress time
        pb$tick()  # update pb
    }
})

Normalize the gene expression matrix with limma

Description

limma_normalize() normalizes an expression matrix using the quantile normalization method provided by the limma package.

Usage

limma_normalize(expressionMatrix)

Arguments

expressionMatrix

A numeric matrix of raw gene expression counts, with genes as rows and cells as columns.

Value

A normalized gene expression matrix after applying limma normalization.

Examples

data(GeneData)
limma_normalize(GeneData)

Fit Linear Model for eQTL Mapping

Description

This function performs linear regression to identify gene–SNP associations based on single-cell expression and genotype data stored in an eQTLObject.

Usage

linearModel(
  eQTLObject,
  geneIDs,
  snpIDs,
  biClassify = FALSE,
  pAdjustMethod = "bonferroni",
  pAdjustThreshold = 0.05,
  logfcThreshold = 0.1
)

Arguments

eQTLObject

An S4 object of class eQTLObject.
geneIDs

Character vector of gene IDs to include in the model fitting.
snpIDs

Character vector of SNP IDs to include in the model fitting.
biClassify

Logical; whether to convert genotype encoding in snpMatrix to 0, 1, and 2. TRUE indicates conversion; FALSE indicates no conversion (default).
pAdjustMethod

Method used for multiple testing correction. One of "bonferroni", "holm", "hochberg", "hommel", or "BH". Default is "bonferroni".
pAdjustThreshold

Only SNP–gene pairs with adjusted p-values below the threshold will be retained. Default is 0.05.
logfcThreshold

The minimum beta coefficient (effect size) required to report a SNP–gene pair as an eQTL.

Value

A data frame of gene–SNP pairs that pass the filtering criteria, including beta coefficients, p-values, adjusted p-values, and group labels.

Examples

data(EQTL_obj)
Gene <- rownames(slot(EQTL_obj, "filterData")$expMat)
SNP <- rownames(slot(EQTL_obj, "filterData")$snpMat)
linearResult <- linearModel(
  eQTLObject = EQTL_obj,
  geneIDs = Gene,
  snpIDs = SNP,
  biClassify = FALSE,
  pAdjustMethod = "bonferroni",
  pAdjustThreshold = 0.05,
  logfcThreshold = 0.025)

Access biclassification information from an eQTLObject.

Description

Access biclassification information from an eQTLObject.

Method to access eQTLObject biclassify information.

Usage

load_biclassify_info(x)

## S4 method for signature 'eQTLObject'
load_biclassify_info(x)

Arguments

x

An eQTLObject.

Value

A character or list containing biclassification information.

biclassify information of eQTLObject.

Examples

data(EQTL_obj)
  load_biclassify_info(EQTL_obj)

Access cell grouping information from an eQTLObject

Description

Access cell grouping information from an eQTLObject

Method to access eQTLObject cell grouping information.

Usage

load_group_info(x)

## S4 method for signature 'eQTLObject'
load_group_info(x)

Arguments

x

An eQTLObject.

Value

A data frame with grouping information.

A data frame with cell group assignments.

Examples

data(EQTL_obj)
  load_group_info(EQTL_obj)

Access species information from an eQTLObject.

Description

Access species information from an eQTLObject.

Method to access eQTLObject species information.

Usage

load_species_info(x)

## S4 method for signature 'eQTLObject'
load_species_info(x)

Arguments

x

An eQTLObject.

Value

A character string indicating the species.

species information of eQTLObject.

Examples

data(EQTL_obj)
  load_species_info(EQTL_obj)

Normalize the gene expression matrix with logNormalize method.

Description

log_normalize() transforms an expression matrix by applying logarithm and scaling operations to normalize data.

Usage

log_normalize(expressionMatrix)

Arguments

expressionMatrix

A numeric matrix of raw gene expression counts, with genes as rows and cells as columns.

Value

A normalized gene expression matrix after applying logNormalize normalization.

Examples

data(GeneData)
log_normalize(GeneData)

Normalize gene expression data.

Description

This function performs normalization of the gene expression matrix stored within an eQTLObject, aiming to remove technical biases and improve comparability across cells. Multiple normalization methods are supported, including "logNormalize", "CPM", "TPM", "DESeq", and "limma".

Usage

normalizeGene(eQTLObject, method = "logNormalize")

Arguments

eQTLObject

An S4 object of class eQTLObject containing the raw gene expression matrix.
method

Character string specifying the normalization method to use. Must be one of "logNormalize", "CPM", "TPM", "DESeq", or "limma". Default is "logNormalize".

Value

An eQTLObject with the normalized gene expression matrix stored in the slot named "normExpMat".

Examples

data(EQTL_obj)
eqtl <- normalizeGene(EQTL_obj, method = "logNormalize")

Customized ggplot2 Theme for Plots

Description

Customized ggplot2 Theme for Plots

Usage

plots_theme_opts()

Value

A ggplot2 theme object for styling plots.

Examples

library(ggplot2)
data <- data.frame(
  x = c("A", "B", "C", "D", "E"),
  y = c(10, 20, 30, 40, 50))
ggplot(data, aes(x, y)) +
  geom_point() +
  plots_theme_opts()

Poisson model fitting the gene expression matrix and genotype matrix.

Description

Poisson model fitting the gene expression matrix and genotype matrix.

Usage

poissonModel(
  eQTLObject,
  geneIDs,
  snpIDs,
  biClassify = FALSE,
  pAdjustMethod = "bonferroni",
  pAdjustThreshold = 0.05,
  logfcThreshold = 0.1
)

Arguments

eQTLObject

An S4 object of class eQTLObject.
geneIDs

Character vector of gene IDs to include in the model fitting.
snpIDs

Character vector of SNP IDs to include in the model fitting.
biClassify

Logical; whether to convert genotype encoding in snpMatrix to 0, 1, and 2. TRUE indicates conversion; FALSE indicates no conversion (default).
pAdjustMethod

Method used for multiple testing correction. One of "bonferroni", "holm", "hochberg", "hommel", or "BH". Default is "bonferroni".
pAdjustThreshold

Only SNP–gene pairs with adjusted p-values below the threshold will be retained. Default is 0.05.
logfcThreshold

The minimum beta coefficient (effect size) required to report a SNP–gene pair as an eQTL.

Value

A data frame of gene–SNP pairs that pass the filtering criteria, including beta coefficients, p-values, adjusted p-values, and group labels.

Examples

data(EQTL_obj)
Gene <- rownames(slot(EQTL_obj, "filterData")$expMat)
SNP <- rownames(slot(EQTL_obj, "filterData")$snpMat)
poissonResult <- poissonModel(
  eQTLObject = EQTL_obj,
  geneIDs = Gene,
  snpIDs = SNP,
  biClassify = FALSE,
  pAdjustMethod = "bonferroni",
  pAdjustThreshold = 0.05,
  logfcThreshold = 0.025
)

Process a matrix to extract a row and convert it to a data frame.

Description

Process a matrix to extract a row and convert it to a data frame.

Usage

process_matrix(id, matrix, name)

Arguments

id

Character string specifying the row name to extract from the matrix.
matrix

A matrix from which the row will be extracted.
name

Character string specifying the column name for the extracted values.

Value

A data frame containing the extracted row and a column with the row names.

Examples

rownames <- c("CNN2", "TIGD2", "DTD2")
colnames <- c("Col1", "Col2", "Col3", "Col4")
matrix_data <- matrix(1:12, nrow = 3, ncol = 4,
  dimnames = list(rownames, colnames))
geneid <- "CNN2"
gene_mat <- process_matrix(geneid, matrix_data, "gene_mat")

Remove outliers from gene expression data and update cell groups.

Description

This function identifies and removes outlier expression values for a specified gene based on the Median Absolute Deviation (MAD) method. It then filters provided genotype-based cell groups, returning only cells with non-outlier expression values.

Usage

remove_outliers(exprsMat, Geneid, A_cells, B_cells, C_cells = NULL)

Arguments

exprsMat

Input gene expression matrix with genes as rows and cells as columns.
Geneid

Character string specifying the gene ID to examine.
A_cells

Character vector of cell names belonging to genotype group A.
B_cells

Character vector of cell names belonging to genotype group B.
C_cells

Optional character vector of cell names belonging to genotype group C; if NULL, function returns two genotype groups.

Value

A named list of filtered cell vectors.

Examples

## Mock expression matrix
set.seed(123)
exprsMat <- matrix(rnorm(200), nrow = 5)
rownames(exprsMat) <- paste0("Gene", 1:nrow(exprsMat))
colnames(exprsMat)  <- paste0("cell", 1:ncol(exprsMat))
A_cells <- colnames(exprsMat)[1:13] # Example A cell list
B_cells <- colnames(exprsMat)[14:26]  # Example B cell list
C_cells <- colnames(exprsMat)[27:40]  # Example C cell list
remove_outliers(exprsMat, "Gene1", A_cells, B_cells, C_cells)

Set filtered data in an eQTLObject.

Description

Set filtered data in an eQTLObject.

Method to set eQTLObject filter data.

Usage

set_filter_data(x, value, name)

## S4 method for signature 'eQTLObject'
set_filter_data(x, value, name)

Arguments

x

An eQTLObject.
value

A matrix to be stored as filtered data.
name

A character string indicating the key under which the matrix is stored in filterData.

Value

An updated eQTLObject.

An updated eQTLObject.

Examples

data(EQTL_obj)
  data123 <- matrix(0, nrow = 3, ncol = 3)
  set_filter_data(EQTL_obj, data123, "expMat")

  data(EQTL_obj)
  data123 <- matrix(0, nrow = 3, ncol = 3)
  set_filter_data(EQTL_obj, data123, "expMat")

Set model specification in an eQTLObject

Description

Set model specification in an eQTLObject

Method to set eQTLObject used model information.

Usage

set_model_info(x, value)

## S4 method for signature 'eQTLObject'
set_model_info(x, value)

Arguments

x

An eQTLObject.
value

A character string specifying the model (e.g., "zinb", "poisson", "linear").

Value

An updated eQTLObject with the new model specification.

An updated eQTLObject.

Examples

data(EQTL_obj)
  useModel <- "zinb"
  set_model_info(EQTL_obj, useModel)

  data(EQTL_obj)
  useModel <- "zinb"
  set_model_info(EQTL_obj, useModel)

Set raw data in an eQTLObject.

Description

Set raw data in an eQTLObject.

Method to set eQTLObject raw data.

Usage

set_raw_data(x, value, name)

## S4 method for signature 'eQTLObject'
set_raw_data(x, value, name)

Arguments

x

An eQTLObject.
value

The raw data.
name

A character string indicating the key under which the matrix is stored in rawData.

Value

eQTLObject.

An updated eQTLObject.

Examples

data(EQTL_obj)
  data123 <- matrix(0, nrow = 3, ncol = 3)
  set_raw_data(EQTL_obj, data123, "rawExpMat")

  data(EQTL_obj)
  data123 <- matrix(0, nrow = 3, ncol = 3)
  set_raw_data(EQTL_obj, data123, "rawExpMat")

Set eQTL results in an eQTLObject.

Description

Set eQTL results in an eQTLObject.

Method to set the result of identifying eQTLs from scRNA-seq data.

Usage

set_result_info(x, value)

## S4 method for signature 'eQTLObject'
set_result_info(x, value)

Arguments

x

An eQTLObject.
value

A data frame where each row corresponds to a SNP–gene pair.

Value

An updated eQTLObject.

An updated eQTLObject.

Examples

data(EQTL_obj)
  result <- data.frame(0, nrow = 3, ncol = 3)
  set_result_info(EQTL_obj, result)

  data(EQTL_obj)
  result <- matrix(0, nrow = 3, ncol = 3)
  set_result_info(EQTL_obj, result)

Normalize the gene expression matrix with TPM

Description

TPM_normalize() scales an expression matrix using Transcripts Per Million (TPM) normalization, applying logarithm and scaling operations to adjust data based on library size.

Usage

TPM_normalize(expressionMatrix)

Arguments

expressionMatrix

A numeric matrix of raw gene expression counts, with genes as rows and cells as columns.

Value

A normalized gene expression matrix after applying TPM normalization.

Examples

data(GeneData)
TPM_normalize(GeneData)

Visualize eQTL results by SNP–gene pair across groups

Description

Visualize eQTL results by SNP–gene pair across groups

Usage

visualizeQTL(
  eQTLObject,
  SNPid,
  Geneid,
  groupName = NULL,
  plottype = "QTLplot",
  removeoutlier = FALSE
)

Arguments

eQTLObject

An S4 object of class eQTLObject.
SNPid

ID of SNP.
Geneid

ID of Gene.
groupName

A character vector specifying one or more cell group names to include in the plot.
plottype

A character string specifying the type of plot to generate. Must be one of "QTLplot", "violin", "boxplot", or "histplot".
removeoutlier

Logical; whether to identify and remove outliers. Default is FALSE.

Value

list

Examples

data(EQTL_obj)
## We have to call the eQTLs firstly using \code{\link{callQTL()}}.
eqtl <- callQTL(eQTLObject = EQTL_obj, useModel = "linear")
visualizeQTL(eQTLObject = eqtl,
SNPid = "1:632647",
Geneid = "RPS27",
groupName = NULL,
plottype = "QTLplot",
removeoutlier = FALSE)

Zinb model fitting the gene expression matrix.

Description

Zinb model fitting the gene expression matrix.

Usage

zinbModel(
  eQTLObject,
  geneIDs,
  snpIDs,
  biClassify = FALSE,
  pAdjustMethod = "bonferroni",
  pAdjustThreshold = 0.05
)

Arguments

eQTLObject

An S4 object of class eQTLObject.
geneIDs

Character vector of gene IDs to include in the model fitting.
snpIDs

Character vector of SNP IDs to include in the model fitting.
biClassify

Logical; whether to convert genotype encoding in snpMatrix to 0, 1, and 2. TRUE indicates conversion; FALSE indicates no conversion (default).
pAdjustMethod

Method used for multiple testing correction. One of "bonferroni", "holm", "hochberg", "hommel", or "BH". Default is "bonferroni".
pAdjustThreshold

Only gene-SNP pairs with adjusted p-values below the threshold will be retained. Default is 0.05.

Value

A data frame of SNP–gene pairs that pass the filtering criteria, including p-values, adjusted p-values, and group labels.

Examples

data(EQTL_obj)
Gene <- rownames(slot(EQTL_obj, 'filterData')$expMat)
SNP <- rownames(slot(EQTL_obj, 'filterData')$snpMat)
zinbResult <- zinbModel(
  eQTLObject = EQTL_obj,
  geneIDs = Gene,
  snpIDs = SNP,
  biClassify = FALSE,
  pAdjustMethod = 'bonferroni',
  pAdjustThreshold = 0.05)