Package: digitalDLSorteR 1.1.2

Diego Mañanes

digitalDLSorteR: Deconvolution of Bulk RNA-Seq Data Based on Deep Learning

Deconvolution of bulk RNA-Seq data using context-specific deconvolution models based on Deep Neural Networks using scRNA-Seq data as input. These models are able to make accurate estimates of the cell composition of bulk RNA-Seq samples from the same context using the advances provided by Deep Learning and the meaningful information provided by scRNA-Seq data. See Torroja and Sanchez-Cabo (2019) <doi:10.3389/fgene.2019.00978> for more details.

Authors:Diego Mañanes [aut, cre], Carlos Torroja [aut], Fatima Sanchez-Cabo [aut]

digitalDLSorteR_1.1.2.tar.gz
digitalDLSorteR_1.1.2.zip(r-4.5)digitalDLSorteR_1.1.2.zip(r-4.4)digitalDLSorteR_1.1.2.zip(r-4.3)
digitalDLSorteR_1.1.2.tgz(r-4.4-any)digitalDLSorteR_1.1.2.tgz(r-4.3-any)
digitalDLSorteR_1.1.2.tar.gz(r-4.5-noble)digitalDLSorteR_1.1.2.tar.gz(r-4.4-noble)
digitalDLSorteR_1.1.2.tgz(r-4.4-emscripten)digitalDLSorteR_1.1.2.tgz(r-4.3-emscripten)
digitalDLSorteR.pdf |digitalDLSorteR.html✨
digitalDLSorteR/json (API)
NEWS

# Install 'digitalDLSorteR' in R:

install.packages('digitalDLSorteR', repos = c('https://diegommcc.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/diegommcc/digitaldlsorter/issues

On CRAN:

deconvolution deep-learning rna-seq single-cell

6.10 score 9 stars 5 scripts 101 downloads 78 exports 161 dependencies

Last updated 23 days agofrom:c40770b26d. Checks:OK: 6 ERROR: 1. Indexed: yes.

Target	Result	Date
Doc / Vignettes	OK	Oct 31 2024
R-4.5-win	OK	Oct 31 2024
R-4.5-linux	ERROR	Oct 31 2024
R-4.4-win	OK	Oct 31 2024
R-4.4-mac	OK	Oct 31 2024
R-4.3-win	OK	Oct 31 2024
R-4.3-mac	OK	Oct 31 2024

Exports:barErrorPlot barPlotCellTypes blandAltmanLehPlot bulk.simul bulk.simul<-calculateEvalMetrics cell.names cell.names<-cell.types cell.types<-corrExpPredPlot createDDLSobject deconv.data deconv.data<-deconv.results deconv.results<-deconvDDLSObj deconvDDLSPretrained DigitalDLSorter DigitalDLSorterDNN distErrorPlot estimateZinbwaveParams features features<-generateBulkCellMatrix getProbMatrix installTFpython interGradientsDL listToDDLS listToDDLSDNN loadDeconvData loadTrainedModelFromH5 method method<-model model<-plotHeatmapGradsAgg plots plots<-plotTrainingHistory preparingToSave prob.cell.types prob.cell.types<-prob.matrix prob.matrix<-ProbMatrixCellTypes project project<-saveRDS saveTrainedModelAsH5 set set.list set.list<-set<-showProbPlot simBulkProfiles simSCProfiles single.cell.real single.cell.real<-single.cell.simul single.cell.simul<-test.deconv.metrics test.deconv.metrics<-test.metrics test.metrics<-test.pred test.pred<-topGradientsCellType trainDDLSModel trained.model trained.model<-training.history training.history<-zinb.params zinb.params<-ZinbParametersModel zinbwave.model zinbwave.model<-

Dependencies:abind annotate AnnotationDbi askpass assorthead backports base64enc beachmat BH Biobase BiocGenerics BiocNeighbors BiocParallel BiocSingular Biostrings bit bit64 blob bluster boot broom cachem car carData cli cluster codetools colorspace config corrplot cowplot cpp11 crayon curl DBI DelayedArray Deriv doBy dplyr dqrng edgeR fansi farver fastmap formatR Formula futile.logger futile.options genefilter generics GenomeInfoDb GenomeInfoDbData GenomicRanges ggplot2 ggpubr ggrepel ggsci ggsignif glue gridExtra grr gtable gtools here httr igraph IRanges irlba isoband jsonlite KEGGREST keras labeling lambda.r lattice lifecycle limma lme4 locfit magrittr MASS Matrix MatrixGenerics MatrixModels matrixStats memoise metapod mgcv microbenchmark mime minqa modelr munsell nlme nloptr nnet numDeriv openssl pbapply pbkrtest pillar pkgconfig plogr plyr png polynom processx ps purrr quantreg R6 rappdirs RColorBrewer Rcpp RcppEigen RcppTOML reshape2 reticulate rlang rprojroot RSQLite rstatix rstudioapi rsvd S4Arrays S4Vectors ScaledMatrix scales scran scuttle SingleCellExperiment sitmo snow softImpute SparseArray SparseM statmod stringi stringr SummarizedExperiment survival sys tensorflow tfautograph tfruns tibble tidyr tidyselect UCSC.utils utf8 vctrs viridisLite whisker withr XML xtable XVector yaml zeallot zinbwave zlibbioc

HDF5 files as back-end

Rendered fromhdf5Backend.Rmdusingknitr::rmarkdownon Oct 31 2024.

Last update: 2024-04-02
Started: 2021-08-10

Keras/TensorFlow installation and configuration

Rendered fromkerasIssues.Rmdusingknitr::rmarkdownon Oct 31 2024.

Last update: 2024-09-29
Started: 2021-08-10

Building new deconvoluion models: deconvolution of colorectal cancer samples

Rendered fromrealModelWorkflow.Rmdusingknitr::rmarkdownon Oct 31 2024.

Last update: 2024-09-30
Started: 2024-04-02

Using pre-trained context-specific deconvolution models

Rendered frompretrainedModels.Rmdusingknitr::rmarkdownon Oct 31 2024.

Last update: 2024-09-30
Started: 2021-08-10

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Generate bar error plots	barErrorPlot
Bar plot of deconvoluted cell type proportions in bulk RNA-Seq samples	barPlotCellTypes barPlotCellTypes,ANY-method barPlotCellTypes,DigitalDLSorter-method
Generate Bland-Altman agreement plots between predicted and expected cell type proportions from test data results	blandAltmanLehPlot
Get and set 'bulk.simul' slot in a 'DigitalDLSorter' object	bulk.simul bulk.simul,DigitalDLSorter-method bulk.simul<- bulk.simul<-,DigitalDLSorter-method
Calculate evaluation metrics for bulk RNA-Seq samples from test data	calculateEvalMetrics
Get and set 'cell.names' slot in a 'ProbMatrixCellTypes' object	cell.names cell.names,ProbMatrixCellTypes-method cell.names<- cell.names<-,ProbMatrixCellTypes-method
Get and set 'cell.types' slot in a 'DigitalDLSorterDNN' object	cell.types cell.types,DigitalDLSorterDNN-method cell.types<- cell.types<-,DigitalDLSorterDNN-method
Generate correlation plots between predicted and expected cell type proportions from test data	corrExpPredPlot
Create a 'DigitalDLSorter' object from single-cell RNA-seq and bulk RNA-seq data	createDDLSobject
Get and set 'deconv.data' slot in a 'DigitalDLSorter' object	deconv.data deconv.data,DigitalDLSorter-method deconv.data<- deconv.data<-,DigitalDLSorter-method
Get and set 'deconv.results' slot in a 'DigitalDLSorter' object	deconv.results deconv.results,DigitalDLSorter-method deconv.results<- deconv.results<-,DigitalDLSorter-method
Deconvolute bulk gene expression samples (bulk RNA-Seq)	deconvDDLSObj
Deconvolute bulk RNA-Seq samples using a pre-trained DigitalDLSorter model	deconvDDLSPretrained
digitalDLSorteR: an R package to deconvolute bulk RNA-Seq samples using single-cell RNA-seq data and neural networks	digitalDLSorteR-package digitalDLSorteR
The DigitalDLSorter Class	DigitalDLSorter DigitalDLSorter-class
The DigitalDLSorterDNN Class	DigitalDLSorterDNN DigitalDLSorterDNN-class
Generate box plots or violin plots to show how the errors are distributed	distErrorPlot
Estimate the parameters of the ZINB-WaVE model to simulate new single-cell RNA-Seq expression profiles	estimateZinbwaveParams
Get and set 'features' slot in a 'DigitalDLSorterDNN' object	features features,DigitalDLSorterDNN-method features<- features<-,DigitalDLSorterDNN-method
Generate training and test cell composition matrices	generateBulkCellMatrix
Getter function for the cell composition matrix	getProbMatrix
Install Python dependencies for digitalDLSorteR	installTFpython
Calculate gradients of predicted cell types/loss function with respect to input features for interpreting trained deconvolution models	interGradientsDL
Transform DigitalDLSorter-like list into an actual DigitalDLSorterDNN object	listToDDLS
Transform DigitalDLSorterDNN-like list into an actual DigitalDLSorterDNN object	listToDDLSDNN
Load data to be deconvoluted into a DigitalDLSorter object	loadDeconvData loadDeconvData,DigitalDLSorter,character-method loadDeconvData,DigitalDLSorter,SummarizedExperiment-method
Load from an HDF5 file a trained Deep Neural Network model into a 'DigitalDLSorter' object	loadTrainedModelFromH5
Get and set 'method' slot in a 'ProbMatrixCellTypes' object	method method,ProbMatrixCellTypes-method method<- method<-,ProbMatrixCellTypes-method
Get and set 'model' slot in a 'DigitalDLSorterDNN' object	model model,DigitalDLSorterDNN-method model<- model<-,DigitalDLSorterDNN-method
Plot a heatmap of gradients of classes / loss function wtih respect to the input	plotHeatmapGradsAgg
Get and set 'plots' slot in a 'ProbMatrixCellTypes' object	plots plots,ProbMatrixCellTypes-method plots<- plots<-,ProbMatrixCellTypes-method
Plot training history of a trained DigitalDLSorter Deep Neural Network model	plotTrainingHistory
Prepare 'DigitalDLSorter' object to be saved as an RDA file	preparingToSave
Get and set 'prob.cell.types' slot in a 'DigitalDLSorter' object	prob.cell.types prob.cell.types,DigitalDLSorter-method prob.cell.types<- prob.cell.types<-,DigitalDLSorter-method
Get and set 'prob.matrix' slot in a 'ProbMatrixCellTypes' object	prob.matrix prob.matrix,ProbMatrixCellTypes-method prob.matrix<- prob.matrix<-,ProbMatrixCellTypes-method
The Class ProbMatrixCellTypes	ProbMatrixCellTypes ProbMatrixCellTypes-class
Get and set 'project' slot in a 'DigitalDLSorter' object	project project,DigitalDLSorter-method project<- project<-,DigitalDLSorter-method
Save 'DigitalDLSorter' objects as RDS files	saveRDS saveRDS,DigitalDLSorter-method saveRDS,DigitalDLSorterDNN-method saveRDS,saveRDS-method
Save a trained 'DigitalDLSorter' Deep Neural Network model to disk as an HDF5 file	saveTrainedModelAsH5
Get and set 'set' slot in a 'ProbMatrixCellTypes' object	set set,ProbMatrixCellTypes-method set<- set<-,ProbMatrixCellTypes-method
Get and set 'set.list' slot in a 'ProbMatrixCellTypes' object	set.list set.list,ProbMatrixCellTypes-method set.list<- set.list<-,ProbMatrixCellTypes-method
Show distribution plots of the cell proportions generated by 'generateBulkCellMatrix'	showProbPlot
Simulate training and test pseudo-bulk RNA-Seq profiles	simBulkProfiles
Simulate new single-cell RNA-Seq expression profiles using the ZINB-WaVE model parameters	simSCProfiles
Get and set 'single.cell.real' slot in a 'DigitalDLSorter' object	single.cell.real single.cell.real,DigitalDLSorter-method single.cell.real<- single.cell.real<-,DigitalDLSorter-method
Get and set 'single.cell.simul' slot in a 'DigitalDLSorter' object	single.cell.simul single.cell.simul,DigitalDLSorter-method single.cell.simul<- single.cell.simul<-,DigitalDLSorter-method
Get and set 'test.deconv.metrics' slot in a 'DigitalDLSorterDNN' object	test.deconv.metrics test.deconv.metrics,DigitalDLSorterDNN-method test.deconv.metrics<- test.deconv.metrics<-,DigitalDLSorterDNN-method
Get and set 'test.metrics' slot in a 'DigitalDLSorterDNN' object	test.metrics test.metrics,DigitalDLSorterDNN-method test.metrics<- test.metrics<-,DigitalDLSorterDNN-method
Get and set 'test.pred' slot in a 'DigitalDLSorterDNN' object	test.pred test.pred,DigitalDLSorterDNN-method test.pred<- test.pred<-,DigitalDLSorterDNN-method
Get top genes with largest/smallest gradients per cell type	topGradientsCellType
Train Deep Neural Network model	trainDDLSModel
Get and set 'trained.model' slot in a 'DigitalDLSorter' object	trained.model trained.model,DigitalDLSorter-method trained.model<- trained.model<-,DigitalDLSorter-method
Get and set 'training.history' slot in a 'DigitalDLSorterDNN' object	training.history training.history,DigitalDLSorterDNN-method training.history<- training.history<-,DigitalDLSorterDNN-method
Get and set 'zinb.params' slot in a 'DigitalDLSorter' object	zinb.params zinb.params,DigitalDLSorter-method zinb.params<- zinb.params<-,DigitalDLSorter-method
The Class ZinbParametersModel	ZinbParametersModel ZinbParametersModel-class
Get and set 'zinbwave.model' slot in a 'ZinbParametersModel' object	zinbwave.model zinbwave.model,ZinbParametersModel-method zinbwave.model<- zinbwave.model<-,ZinbParametersModel-method