Package 'ADImpute' reference manual

Title:	Adaptive Dropout Imputer (ADImpute)
Description:	Single-cell RNA sequencing (scRNA-seq) methods are typically unable to quantify the expression levels of all genes in a cell, creating a need for the computational prediction of missing values (‘dropout imputation’). Most existing dropout imputation methods are limited in the sense that they exclusively use the scRNA-seq dataset at hand and do not exploit external gene-gene relationship information. Here we propose two novel methods: a gene regulatory network-based approach using gene-gene relationships learnt from external data and a baseline approach corresponding to a sample-wide average. ADImpute can implement these novel methods and also combine them with existing imputation methods (currently supported: DrImpute, SAVER). ADImpute can learn the best performing method per gene and combine the results from different methods into an ensemble.
Authors:	Ana Carolina Leote [cre, aut]
Maintainer:	Ana Carolina Leote <[email protected]>
License:	GPL-3 + file LICENSE
Version:	1.17.0
Built:	2024-10-30 03:25:46 UTC
Source:	https://github.com/bioc/ADImpute

Data trimming

Description

ArrangeData finds common genes to the network and provided data and limits both datasets to these

Usage

ArrangeData(data, net.coef = NULL)
ArrangeData(data, net.coef = NULL)

Arguments

`data`	matrix with entries equal to zero to be imputed (genes as rows and samples as columns)
`net.coef`	matrix; object containing network coefficients

Value

list; data matrix, network coefficients matrix and intercept for genes common between the data matrix and the network

Data centering

Description

CenterData centers expression of each gene at 0

Usage

CenterData(data)
CenterData(data)

Arguments

data

matrix of gene expression to be centered row-wise (genes as rows and samples as columns)

Value

list; row-wise centers and centered data

Method choice per gene

Description

ChooseMethod determines the method for dropout imputation based on performance on each gene in training data

Usage

ChooseMethod(real, masked, imputed, write.to.file = TRUE)
ChooseMethod(real, masked, imputed, write.to.file = TRUE)

Arguments

`real`	matrix; original gene expression data, i.e. before masking (genes as rows and samples as columns)
`masked`	matrix, logical indicating which entries were masked (genes as rows and samples as columns)
`imputed`	list; list of matrices with imputation results for all considered methods
`write.to.file`	logical; should the output be written to a file?

Details

The imputed values are compared to the real ones for every masked entry in real. The Mean Squared Error is computed for all masked entries per gene and the method with the best performance is chosen for each gene.

Value

character; best performing method in the training set for each gene

Combine imputation methods

Description

Combine imputation methods

Usage

Combine(data, imputed, method.choice, write = FALSE)
Combine(data, imputed, method.choice, write = FALSE)

Arguments

`data`	matrix with entries equal to zero to be imputed, already normalized (genes as rows and samples as columns)
`imputed`	list; list of matrices with imputation results for all considered methods
`method.choice`	named character; vector with the best performing method per gene
`write`	logical; should a file with the imputation results be written?

Details

Combines imputation results from all methods according to training results provided in method.choice

Value

matrix; imputation results combining the best performing method per gene

Computation of MSE per gene

Description

ComputeMSEGenewise computes the MSE of dropout imputation for a given gene.

Usage

ComputeMSEGenewise(real, masked, imputed, baseline)
ComputeMSEGenewise(real, masked, imputed, baseline)

Arguments

`real`	numeric; vector of original expression of a given gene (before masking)
`masked`	logical; vector indicating which entries were masked for a given gene
`imputed`	matrix; imputation results for a given imputation method
`baseline`	logical; is this baseline imputation?

Value

MSE of all imputations indicated by masked

Argument check

Description

CreateArgCheck creates tests for argument correctness.

Usage

CreateArgCheck(missing = NULL, match = NULL, acceptable = NULL,
null = NULL)
CreateArgCheck(missing = NULL, match = NULL, acceptable = NULL,
null = NULL)

Arguments

`missing`	named list; logical. Name corresponds to variable name, and corresponding entry to whether it was missing from the function call.
`match`	named list. Name corresponds to variable name, and corresponding entry to its value.
`acceptable`	named list. Name corresponds to variable name, and corresponding entry to its acceptable values.
`null`	named list; logical. Name corresponds to variable name, and corresponding entry to whether it was NULL in the function call.

Value

argument check object.

Preparation of training data for method evaluation

Description

CreateTrainingData selects a subset of cells to use as training set and sets a portion (mask) of the non-zero entries in each row of the subset to zero

Usage

CreateTrainData(data, train.ratio = .7, train.only = TRUE, mask = .1,
write = FALSE)
CreateTrainData(data, train.ratio = .7, train.only = TRUE, mask = .1,
write = FALSE)

Arguments

`data`	matrix; raw counts (genes as rows and samples as columns)
`train.ratio`	numeric; ratio of the samples to be used for training
`train.only`	logical; if TRUE define only a training dataset, if FALSE writes both training and validation sets (defaults to TRUE)
`mask`	numeric; ratio of total non-zero samples to be masked per gene (defaults to .1)
`write`	logical; should the output be written to a file?

Value

list with resulting matrix after subsetting and after masking

Data check (matrix)

Description

DataCheck_Matrix tests for potential format and storage issues with matrices. Helper function to ADImpute.

Usage

DataCheck_Matrix(data)
DataCheck_Matrix(data)

Arguments

data

data object to check

Value

data object with needed adjustments

Data check (network)

Description

DataCheck_Network tests for potential format and storage issues with the network coefficient matrix. Helper function to ADImpute.

Usage

DataCheck_Network(network)
DataCheck_Network(network)

Arguments

network

data object containing matrix coefficients

Value

network data object with needed adjustments

Data check (SingleCellExperiment)

Description

DataCheck_SingleCellExperiment tests for existence of the appropriate assays in sce. Helper function to ADImpute.

Usage

DataCheck_SingleCellExperiment(sce, normalized = TRUE)
DataCheck_SingleCellExperiment(sce, normalized = TRUE)

Arguments

`sce`	SingleCellExperiment; data for normalization or imputation
`normalized`	logical; is the data expected to be normalized?

Value

NULL object.

Data check (transcript length)

Description

DataCheck_TrLength tests for potential format and storage issues with the object encoding transcript length, for e.g. TPM normalization. Helper function to ADImpute.

Usage

DataCheck_TrLength(trlength)
DataCheck_TrLength(trlength)

Arguments

trlength

data object containing transcript length information

Value

transcript length object with needed adjustments

Small dataset for example purposes

Description

A small dataset to use on vignettes and examples (50 cells).

Usage

demo_data
demo_data

Format

matrix; a subset of the Grun pancreas dataset, obtained with the scRNAseq R package, to use in the vignette and examples.

References

Grun D et al. (2016). De novo prediction of stem cell identity using single-cell transcriptome data. Cell Stem Cell 19(2), 266-277.

Small regulatory network for example purposes

Description

Subset of the Gene Regulatory Network used by ADImpute's Network imputation method.

Usage

demo_net
demo_net

Format

matrix; subset of the Gene Regulatory Network installed along with ADImpute.

Small dataset for example purposes

Description

A small dataset to use on vignettes and examples (50 cells).

Usage

demo_sce
demo_sce

Format

SingleCellExperiment; a subset of the Grun pancreas dataset, obtained with the scRNAseq R package, to use in the vignette and examples.

References

Grun D et al. (2016). De novo prediction of stem cell identity using single-cell transcriptome data. Cell Stem Cell 19(2), 266-277.

Imputation method evaluation on training set

Description

EvaluateMethods returns the best-performing imputation method for each gene in the dataset

Usage

EvaluateMethods(data, sce = NULL, do = c('Baseline', 'DrImpute',
'Network'), write = FALSE, train.ratio = .7, train.only = TRUE,
mask.ratio = .1, outdir = getwd(), scale = 1, pseudo.count = 1,
labels = NULL, cell.clusters = 2, drop_thre = NULL, type = 'count',
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
net.coef = ADImpute::network.coefficients, net.implementation = 'iteration',
tr.length = ADImpute::transcript_length, bulk = NULL, ...)
EvaluateMethods(data, sce = NULL, do = c('Baseline', 'DrImpute',
'Network'), write = FALSE, train.ratio = .7, train.only = TRUE,
mask.ratio = .1, outdir = getwd(), scale = 1, pseudo.count = 1,
labels = NULL, cell.clusters = 2, drop_thre = NULL, type = 'count',
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
net.coef = ADImpute::network.coefficients, net.implementation = 'iteration',
tr.length = ADImpute::transcript_length, bulk = NULL, ...)

Arguments

`data`	matrix; normalized counts, not logged (genes as rows and samples as columns)
`sce`	SingleCellExperiment; normalized counts and associated metadata.
`do`	character; choice of methods to be used for imputation. Currently supported methods are `'Baseline'`, `'DrImpute'` and `'Network'`. Not case-sensitive. Can include one or more methods. Non- supported methods will be ignored.
`write`	logical; write intermediary and imputed objects to files?
`train.ratio`	numeric; ratio of samples to be used for training
`train.only`	logical; if TRUE define only a training dataset, if FALSE writes and returns both training and validation sets (defaults to TRUE)
`mask.ratio`	numeric; ratio of samples to be masked per gene
`outdir`	character; path to directory where output files are written. Defaults to working directory
`scale`	integer; scaling factor to divide all expression levels by (defaults to 1)
`pseudo.count`	integer; pseudo-count to be added to expression levels to avoid log(0) (defaults to 1)
`labels`	character; vector specifying the cell type of each column of `data`
`cell.clusters`	integer; number of cell subpopulations
`drop_thre`	numeric; between 0 and 1 specifying the threshold to determine dropout values
`type`	A character specifying the type of values in the expression matrix. Can be 'count' or 'TPM'
`cores`	integer; number of cores used for paralell computation
`BPPARAM`	parallel back-end to be used during parallel computation. See `BiocParallelParam-class`.
`net.coef`	matrix; network coefficients. Please provide if you don't want to use ADImpute's network model. Must contain one first column 'O' acconting for the intercept of the model and otherwise be an adjacency matrix with hgnc_symbols in rows and columns. Doesn't have to be squared. See `ADImpute::demo_net` for a small example.
`net.implementation`	character; either 'iteration', for an iterative solution, or 'pseudoinv', to use Moore-Penrose pseudo-inversion as a solution. 'pseudoinv' is not advised for big data.
`tr.length`	matrix with at least 2 columns: 'hgnc_symbol' and 'transcript_length'
`bulk`	vector of reference bulk RNA-seq, if available (average across samples)
`...`	additional parameters to pass to network-based imputation

Details

For each gene, a fraction (mask.ratio) of the quantified expression values are set to zero and imputed according to 3 different methods: scImpute, baseline (average gene expression across all cells) or a network-based method. The imputation error is computed for each of the values in the original dataset that was set to 0, for each method. The method resulting in a lowest imputation error for each gene is chosen.

Value

if sce is provided: returns a SingleCellExperiment with the best performing method per gene stored as row-features. Access via SingleCellExperiment::int_elementMetadata(sce)$ADImpute$methods.
if sce is not provided: returns a character with the best performing method in the training set for each gene

Examples

# Normalize demo data
norm_data <- NormalizeRPM(ADImpute::demo_data)
method_choice <- EvaluateMethods(norm_data, do = c('Baseline','DrImpute'),
cores = 2)

# Normalize demo data
norm_data <- NormalizeRPM(ADImpute::demo_data)
method_choice <- EvaluateMethods(norm_data, do = c('Baseline','DrImpute'),
cores = 2)

Get dropout probabilities

Description

GetDropoutProbabilities computes dropout probabilities (probability of being a dropout that should be imputed rather than a true biological zero) using an adaptation of scImpute's approach

Usage

GetDropoutProbabilities(data, thre, cell.clusters, labels = NULL,
type = 'count', cores, BPPARAM, genelen = ADImpute::transcript_length)
GetDropoutProbabilities(data, thre, cell.clusters, labels = NULL,
type = 'count', cores, BPPARAM, genelen = ADImpute::transcript_length)

Arguments

`data`	matrix; original data before imputation
`thre`	numeric; probability threshold to classify entries as biological zeros
`cell.clusters`	integer; number of cell subpopulations
`labels`	character; vector specifying the cell type of each column of `data`
`type`	A character specifying the type of values in the expression matrix. Can be 'count' or 'TPM'
`cores`	integer; number of cores used for paralell computation
`BPPARAM`	parallel back-end to be used during parallel computation. See `BiocParallelParam-class`.
`genelen`	matrix with at least 2 columns: 'hgnc_symbol' and 'transcript_length'

Details

This function follows scImpute's model to distinguish between true biological zeros and dropouts, and is based on adapted code from the scImpute R package.

Value

matrix with same dimensions as data containing the dropout probabilities for the corresponding entries

Get dropout probabilities

Description

GetDropoutProbabilities computes dropout probabilities (probability of being a dropout that should be imputed rather than a true biological zero) using an adaptation of scImpute's approach

Usage

HandleBiologicalZeros(data, imputed, thre = 0.5, cell.clusters,
labels = NULL, type = 'count', cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
genelen = ADImpute::transcript_length, prob.mat = NULL)
HandleBiologicalZeros(data, imputed, thre = 0.5, cell.clusters,
labels = NULL, type = 'count', cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
genelen = ADImpute::transcript_length, prob.mat = NULL)

Arguments

`data`	matrix; original data before imputation
`imputed`	list; imputation results for considered methods
`thre`	numeric; between 0 and 1 specifying the threshold to determine dropout values
`cell.clusters`	integer; number of cell subpopulations
`labels`	character; vector specifying the cell type of each column of `data`
`type`	A character specifying the type of values in the expression matrix. Can be 'count' or 'TPM'
`cores`	integer; number of cores used for paralell computation
`BPPARAM`	parallel back-end to be used during parallel computation. See `BiocParallelParam-class`.
`genelen`	matrix with at least 2 columns: 'hgnc_symbol' and 'transcript_length'
`prob.mat`	matrix with same dimensions as `data` containing the dropout probabilities for the corresponding entries

Details

This function follows scImpute's model to distinguish between true biological zeros and dropouts, and is based on adapted code from the scImpute R package.

Value

list with 2 components: zerofiltered, a list equivalent to imputed but with entries of imputed likely biological zeros set back to zero, and dropoutprobabilities matrix with same dimensions as data containing the dropout probabilities for the corresponding entries

Dropout imputation using different methods

Description

Impute performs dropout imputation on normalized data, based on the choice of imputation methods.

Usage

Impute(data, sce = NULL, do = 'Ensemble', write = FALSE,
outdir = getwd(), method.choice = NULL, scale = 1, pseudo.count = 1,
labels = NULL, cell.clusters = 2, drop_thre = NULL, type = 'count',
tr.length = ADImpute::transcript_length,
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
net.coef = ADImpute::network.coefficients, net.implementation = 'iteration',
bulk = NULL, true.zero.thr = NULL, prob.mat = NULL, ...)
Impute(data, sce = NULL, do = 'Ensemble', write = FALSE,
outdir = getwd(), method.choice = NULL, scale = 1, pseudo.count = 1,
labels = NULL, cell.clusters = 2, drop_thre = NULL, type = 'count',
tr.length = ADImpute::transcript_length,
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
net.coef = ADImpute::network.coefficients, net.implementation = 'iteration',
bulk = NULL, true.zero.thr = NULL, prob.mat = NULL, ...)

Arguments

`data`	matrix; raw counts (genes as rows and samples as columns)
`sce`	SingleCellExperiment; normalized counts and associated metadata.
`do`	character; choice of methods to be used for imputation. Currently supported methods are `'Baseline'`, `'DrImpute'`, `'Network'`, and `'Ensemble'`. Defaults to `'Ensemble'`. Not case-sensitive. Can include one or more methods. Non-supported methods will be ignored.
`write`	logical; write intermediary and imputed objects to files?
`outdir`	character; path to directory where output files are written. Defaults to working directory
`method.choice`	character; best performing method in training data for each gene
`scale`	integer; scaling factor to divide all expression levels by (defaults to 1)
`pseudo.count`	integer; pseudo-count to be added to expression levels to avoid log(0) (defaults to 1)
`labels`	character; vector specifying the cell type of each column of `data`
`cell.clusters`	integer; number of cell subpopulations
`drop_thre`	numeric; between 0 and 1 specifying the threshold to determine dropout values
`type`	A character specifying the type of values in the expression matrix. Can be 'count' or 'TPM'
`tr.length`	matrix with at least 2 columns: 'hgnc_symbol' and 'transcript_length'
`cores`	integer; number of cores used for paralell computation
`BPPARAM`	parallel back-end to be used during parallel computation. See `BiocParallelParam-class`.
`net.coef`	matrix; network coefficients. Please provide if you don't want to use ADImpute's network model. Must contain one first column 'O' acconting for the intercept of the model and otherwise be an adjacency matrix with hgnc_symbols in rows and columns. Doesn't have to be squared. See `ADImpute::demo_net` for a small example.
`net.implementation`	character; either 'iteration', for an iterative solution, or 'pseudoinv', to use Moore-Penrose pseudo-inversion as a solution. 'pseudoinv' is not advised for big data.
`bulk`	vector of reference bulk RNA-seq, if available (average across samples)
`true.zero.thr`	if set to NULL (default), no true zero estimation is performed. Set to numeric value between 0 and 1 for estimation. Value corresponds to the threshold used to determine true zeros: if the probability of dropout is lower than `true.zero.thr`, the imputed entries are set to zero.
`prob.mat`	matrix of the same size as data, filled with the dropout probabilities for each gene in each cell
`...`	additional parameters to pass to network-based imputation

Details

Values that are 0 in data are imputed according to the best-performing methods indicated in method.choice. Currently supported methods are:

Baseline: imputation with average expression across all cells in the dataset. See ImputeBaseline.
Previously published approaches: DrImpute and SAVER.
Network: leverages information from a gene regulatory network to predicted expression of genes that are not quantified based on quantified interacting genes, in the same cell. See ImputeNetwork.
Ensemble: is based on results on a training subset of the data at hand, indicating which method best predicts the expression of each gene. These results are supplied via method.choice. Applies the imputation results of the best performing method to the zero entries of each gene.

If 'Ensemble' is included in do, method.choice has to be provided (use output from EvaluateMethods()). Impute can create a directory imputation containing the imputation results of all methods in do. If true.zero.thr is set, dropout probabilities are computed using scImpute's framework. Expression values with dropout probabilities below true.zero.thr will be set back to 0 if imputed, as they likely correspond to true biological zeros (genes not expressed in cell) rather than technical dropouts (genes expressed but not captured). If sce is set, imputed values by the different methods are added as new assays to sce. Each assay corresponds to one imputation method. If true.zero.thr is set, only the values after filtering for biological zeros will be added. This is different from the output if sce is not set, where the original values before filtering and the dropout probability matrix are returned.

Value

if sce is not set: returns a list of imputation results (normalized, log-transformed) for all selected methods in do. If true.zero.thr is defined, returns a list of 3 elements: 1) a list, imputations, containing the direct imputation results from each method; 2) a list, zerofiltered, containing the results of imputation in imputations after setting biological zeros back to zero; 3) a matrix, dropoutprobabilities, containing the dropout probability matrix used to set biological zeros.
if sce is set: returns a SingleCellExperiment with new assays, each corresponding to one of the imputation methods applied. If true.zero.thr is defined, the assays will contain the results after imputation and setting biological zeros back to zero.

Examples

# Normalize demo data
norm_data <- NormalizeRPM(demo_data)
# Impute with particular method(s)
imputed_data <- Impute(do = 'Network', data = norm_data[,1:10],
net.coef = ADImpute::demo_net)
imputed_data <- Impute(do = 'Network', data = norm_data[,1:10],
net.implementation = 'pseudoinv', net.coef = ADImpute::demo_net)

# Normalize demo data
norm_data <- NormalizeRPM(demo_data)
# Impute with particular method(s)
imputed_data <- Impute(do = 'Network', data = norm_data[,1:10],
net.coef = ADImpute::demo_net)
imputed_data <- Impute(do = 'Network', data = norm_data[,1:10],
net.implementation = 'pseudoinv', net.coef = ADImpute::demo_net)

Impute using average expression across all cells

Description

ImputeBaseline imputes dropouts using gene averages across cells. Zero values are excluded from the mean computation.

Usage

ImputeBaseline(data, write = FALSE, ...)
ImputeBaseline(data, write = FALSE, ...)

Arguments

`data`	matrix with entries equal to zero to be imputed, normalized and log2-transformed (genes as rows and samples as columns)
`write`	logical; should a file with the imputation results be written?
`...`	additional arguments to `saveRDS`

Value

matrix; imputation results considering the average expression values of genes

Use DrImpute

Description

ImputeDrImpute uses the DrImpute package for dropout imputation

Usage

ImputeDrImpute(data, write = FALSE)
ImputeDrImpute(data, write = FALSE)

Arguments

`data`	matrix with entries equal to zero to be imputed, normalized and log2-transformed (genes as rows and samples as columns)
`write`	logical; should a file with the imputation results be written?

Value

matrix; imputation results from DrImpute

Network-based parallel imputation

Description

ImputeNetParallel implements network-based imputation in parallel

Usage

ImputeNetParallel(drop.mat, arranged, cores =
BiocParallel::bpworkers(BPPARAM), type = 'iteration', max.iter = 50,
BPPARAM = BiocParallel::SnowParam(type = "SOCK"))
#'
ImputeNetParallel(drop.mat, arranged, cores =
BiocParallel::bpworkers(BPPARAM), type = 'iteration', max.iter = 50,
BPPARAM = BiocParallel::SnowParam(type = "SOCK"))
#'

Arguments

`drop.mat`	matrix, logical; dropout entries in the data matrix (genes as rows and samples as columns)
`arranged`	list; output of `ArrangeData`
`cores`	integer; number of cores used for paralell computation
`type`	character; either 'iteration', for an iterative solution, or 'pseudoinv', to use Moore-Penrose pseudo-inversion as a solution.
`max.iter`	numeric; maximum number of iterations for network imputation. Set to -1 to remove limit (not recommended)
`BPPARAM`	parallel back-end to be used during parallel computation. See `BiocParallelParam-class`.

Value

matrix; imputation results incorporating network information

Network-based imputation

Description

Network-based imputation

Usage

ImputeNetwork(data, net.coef = NULL,
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
type = 'iteration', write = FALSE, ...)
ImputeNetwork(data, net.coef = NULL,
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
type = 'iteration', write = FALSE, ...)

Arguments

`data`	matrix with entries equal to zero to be imputed, normalized and log2-transformed (genes as rows and samples as columns)
`net.coef`	matrix; network coefficients.
`cores`	integer; number of cores to use
`BPPARAM`	parallel back-end to be used during parallel computation. See `BiocParallelParam-class`.
`type`	character; either 'iteration', for an iterative solution, or 'pseudoinv', to use Moore-Penrose pseudo-inversion as a solution.
`write`	logical; should a file with the imputation results be written?
`...`	additional arguments to `ImputeNetParallel`

Details

Imputes dropouts using a gene regulatory network trained on external data, as provided in net.coef. Dropout expression values are estimated from the expression of their predictor genes and the network coefficients.

Value

matrix; imputation results incorporating network information

Helper function to PseudoInverseSolution_percell

Description

ImputeNPDropouts computes the non-dropout- dependent solution of network imputation for each cell

Usage

ImputeNPDropouts(net, expr)
ImputeNPDropouts(net, expr)

Arguments

`net`	matrix, logical; network coefficients for all dropout (to be imputed) genes that are predictive of the expression of other dropout genes
`expr`	numeric; vector of gene expression for all genes in the cell at hand

Value

vector; imputation results for the non-dropout-dependent genes

Helper function to PseudoInverseSolution_percell

Description

ImputePredictiveDropouts applies Moore-Penrose pseudo-inversion to compute the dropout-dependent solution of network imputation for each cell

Usage

ImputePredictiveDropouts(net, thr = 0.01, expr)
ImputePredictiveDropouts(net, thr = 0.01, expr)

Arguments

`net`	matrix, logical; network coefficients for all dropout (to be imputed) genes that are predictive of the expression of other dropout genes
`thr`	numeric; tolerance threshold to detect zero singular values
`expr`	numeric; vector of gene expression for all genes in the cell at hand

Value

vector; imputation results for the dropout-dependent genes

Use SAVER

Description

ImputeSAVER uses the SAVER package for dropout imputation

Usage

ImputeSAVER(data, cores, try.mean = FALSE, write = FALSE)
ImputeSAVER(data, cores, try.mean = FALSE, write = FALSE)

Arguments

`data`	matrix with entries equal to zero to be imputed, normalized (genes as rows and samples as columns)
`cores`	integer; number of cores to use
`try.mean`	logical; whether to additionally use mean gene expression as prediction
`write`	logical; should a file with the imputation results be written?

Value

matrix; imputation results from SAVER

Masking of entries for performance evaluation

Description

MaskData sets a portion (mask) of the non-zero entries of each row of data to zero

Usage

MaskData(data, write.to.file = FALSE, mask = .1)
MaskData(data, write.to.file = FALSE, mask = .1)

Arguments

`data`	matrix; raw counts (genes as rows and samples as columns)
`write.to.file`	logical; should the output be written to a file?
`mask`	numeric; ratio of total non-zero samples to be masked per gene (defaults to .1)

Details

Sets a portion (mask) of the non-zero entries of each row of data to zero. Result is written to filename.

Value

matrix containing masked raw counts (genes as rows and samples as columns)

Helper mask function

Description

Helper mask function, per feature.

Usage

MaskerPerGene(x, rowmask)
MaskerPerGene(x, rowmask)

Arguments

`x`	logical; data to mask
`rowmask`	numeric; number of samples to be masked per gene

Value

logical containing positions to mask

Transcriptome wide gene regulatory network

Description

Gene Regulatory Network used by ADImpute's Network imputation method. First column, O, corresponds to the intercept of a gene- -specific predicion model. The remaining rows and columns correspond to the adjacency matrix of the inferred network, where rows are target genes and columns are predictors. Genes are identified by their hgnc_symbol.

Usage

network.coefficients
network.coefficients

Format

dgCMatrix

RPM normalization

Description

NormalizeRPM performs RPM normalization, with possibility to log the result

Usage

NormalizeRPM(data, sce = NULL, log = FALSE, scale = 1,
pseudo.count = 1)
NormalizeRPM(data, sce = NULL, log = FALSE, scale = 1,
pseudo.count = 1)

Arguments

`data`	matrix; raw data (genes as rows and samples as columns)
`sce`	SingleCellExperiment; raw data
`log`	logical; log RPMs?
`scale`	integer; scale factor to divide RPMs by
`pseudo.count`	numeric; if `log = TRUE`, value to add to RPMs in order to avoid taking `log(0)`

Value

matrix; library size normalized data

Examples

demo <- NormalizeRPM(ADImpute::demo_data)

demo <- NormalizeRPM(ADImpute::demo_data)

TPM normalization

Description

NormalizeTPM performs TPM normalization, with possibility to log the result

Usage

NormalizeTPM(data, sce = NULL, tr_length = NULL, log = FALSE,
scale = 1, pseudo.count = 1)
NormalizeTPM(data, sce = NULL, tr_length = NULL, log = FALSE,
scale = 1, pseudo.count = 1)

Arguments

`data`	matrix; raw data (genes as rows and samples as columns)
`sce`	SingleCellExperiment; raw data
`tr_length`	data.frame with at least 2 columns: 'hgnc_symbol' and 'transcript_length'
`log`	logical; log TPMs?
`scale`	integer; scale factor to divide TPMs by
`pseudo.count`	numeric; if `log = T`, value to add to TPMs in order to avoid taking `log(0)`

Details

Gene length is estimated as the median of the lengths of all transcripts for each gene, as obtained from biomaRt. Genes for which length information cannot be found in biomaRt are dropped.

Value

matrix; normalized data (for transcript length and library size)

Examples

demo <- NormalizeTPM(ADImpute::demo_data)

demo <- NormalizeTPM(ADImpute::demo_data)

Network-based parallel imputation - Moore-Penrose pseudoinversion

Description

PseudoInverseSolution_percell applies Moore-Penrose pseudo-inversion to compute the solution of network imputation for each cell

Usage

PseudoInverseSolution_percell(expr, net, drop_ind, thr = 0.01)
PseudoInverseSolution_percell(expr, net, drop_ind, thr = 0.01)

Arguments

`expr`	numeric; expression vector for cell at hand
`net`	matrix; network coefficients
`drop_ind`	logical; dropout entries in the cell at hand
`thr`	numeric; tolerance threshold to detect zero singular values

Value

matrix; imputation results incorporating network information

Data read

Description

ReadData reads data from raw input file (.txt or .csv)

Usage

ReadData(path, ...)
ReadData(path, ...)

Arguments

`path`	character; path to input file
`...`	additional arguments to `data.table::fread()`

Value

matrix; raw counts (genes as rows and samples as columns)

Wrapper for return of EvaluateMethods()

Description

ReturnChoice Adjusts the output of EvaluateMethods to a character vector or a SingleCellExperiment object. Helper function to ADImpute.

Usage

ReturnChoice(sce, choice)
ReturnChoice(sce, choice)

Arguments

`sce`	SingleCellExperiment; a SingleCellExperiment object if available; NULL otherwise
`choice`	character; best performing method in the training set for each gene

Value

if sce is provided: returns a SingleCellExperiment with the best performing method per gene stored as row-features. Access via SingleCellExperiment::int_elementMetadata(sce)$ADImpute$methods.
if sce is not provided: returns a character with the best performing method in the training set for each gene

Wrapper for return of Impute()

Description

ReturnOut Adjusts the output of Impute to a list of matrices or a SingleCellExperiment object. Helper function to ADImpute.

Usage

ReturnOut(result, sce)
ReturnOut(result, sce)

Arguments

`result`	list; imputation result
`sce`	SingleCellExperiment; a SingleCellExperiment object if available; NULL otherwise

Value

imputation results. A SingleCellExperiment if !is.null(sce), or a list with imputed results in matrix format otherwise.

Set biological zeros

Description

SetBiologicalZeros sets some of the entries back to zero after dropout imputation, as they likely correspond to true biological zeros (genes not expressed in given cell)

Usage

SetBiologicalZeros(imputation, drop_probs, thre = .2, was_zero)
SetBiologicalZeros(imputation, drop_probs, thre = .2, was_zero)

Arguments

`imputation`	matrix; imputed values
`drop_probs`	matrix; dropout probabilities for each entry in `imputation`. 0 means certain biological zero, while 1 means certain dropout to be imputed
`thre`	numeric; probability threshold to classify entries as biological zeros
`was_zero`	matrix; logical matrix: was the corresponding entry of `imputation` originally a zero?

Details

Entries which were originally zero and have dropout probability below thre are considered biological zeros and, if they were imputed, are set back to 0.

Value

matrix containing likely biological zeros set back to 0.

Selection of samples for training

Description

SplitData selects a portion (ratio) of samples (columns in data) to be used as training set

Usage

SplitData(data, ratio = .7, write.to.file = FALSE, train.only = TRUE)
SplitData(data, ratio = .7, write.to.file = FALSE, train.only = TRUE)

Arguments

`data`	matrix; raw counts (genes as rows and samples as columns)
`ratio`	numeric; ratio of the samples to be used for training
`write.to.file`	logical; should the output be written to a file?
`train.only`	logical; if TRUE define only a training dataset, if FALSE writes both training and validation sets (defaults to TRUE)

Details

Selects a portion (ratio) of samples (columns in data) to be used as training set and writes to file 'training_raw.txt'.

Value

matrix containing raw counts (genes as rows and samples as columns)

Table for transcript length calculations

Description

A data.frame to be used for transcript length computations. May be necessary upon TPM normalization, or as input to scImpute. All data was retrieved from biomaRt.

Usage

transcript_length
transcript_length

Format

A data.frame with 2 columns:

hgnc_symbol: Gene symbol identifier
transcript length: Length of transcript

Write csv file

Description

WriteCSV writes data to a comma-delimited output file

Usage

WriteCSV(object, file)
WriteCSV(object, file)

Arguments

`object`	R object to write
`file`	character; path to output file

Value

Returns NULL

Examples

file <- tempfile()
WriteCSV(iris, file = file)

file <- tempfile()
WriteCSV(iris, file = file)

Write txt file

Description

WriteTXT writes data to a tab-delimited output file

Usage

WriteTXT(object, file)
WriteTXT(object, file)

Arguments

`object`	R object to write
`file`	character; path to output file

Value

Returns NULL

Examples

file <- tempfile()
WriteTXT(iris, file = file)

file <- tempfile()
WriteTXT(iris, file = file)

Package 'ADImpute'

Help Index

Data trimming

Description

Usage

Arguments

Value

Data centering

Description

Usage

Arguments

Value

Argument check to Impute()

Description

Usage

Arguments

Value

Method choice per gene

Description

Usage

Arguments

Details

Value

See Also

Combine imputation methods

Description

Usage

Arguments

Details

Value

Computation of MSE per gene

Description

Usage

Arguments

Value

Argument check

Description

Usage

Arguments

Value

Preparation of training data for method evaluation

Description

Usage

Arguments

Value

Data check (matrix)

Description

Usage

Arguments

Value

Data check (network)

Description

Usage

Arguments

Value

Data check (SingleCellExperiment)

Description

Usage

Arguments

Value

Data check (transcript length)

Description

Usage

Arguments

Value

Small dataset for example purposes

Description

Usage

Format

References

Small regulatory network for example purposes

Description

Usage

Format

Small dataset for example purposes

Description

Usage

Format

References

Imputation method evaluation on training set