Package: scGPS 1.27.0

Quan Nguyen

scGPS: A complete analysis of single cell subpopulations, from identifying subpopulations to analysing their relationship (scGPS = single cell Global Predictions of Subpopulation)

The package implements two main algorithms to answer two key questions: a SCORE (Stable Clustering at Optimal REsolution) to find subpopulations, followed by scGPS to investigate the relationships between subpopulations.

Authors:Quan Nguyen [aut, cre], Michael Thompson [aut], Anne Senabouth [aut]

scGPS_1.27.0.tar.gz
scGPS_1.27.0.zip(r-4.7)scGPS_1.27.0.zip(r-4.6)scGPS_1.27.0.zip(r-4.5)
scGPS_1.27.0.tgz(r-4.6-x86_64)scGPS_1.27.0.tgz(r-4.6-arm64)scGPS_1.27.0.tgz(r-4.5-x86_64)scGPS_1.27.0.tgz(r-4.5-arm64)
scGPS_1.27.0.tar.gz(r-4.7-arm64)scGPS_1.27.0.tar.gz(r-4.7-x86_64)scGPS_1.27.0.tar.gz(r-4.6-arm64)scGPS_1.27.0.tar.gz(r-4.6-x86_64)
scGPS_1.27.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
scGPS/json (API)

# Install 'scGPS' in R:

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

Bug tracker:https://github.com/imb-computational-genomics-lab/scgps/issues

Uses libs:

openblas– Optimized BLAS
c++– GNU Standard C++ Library v3

Datasets:

day_2_cardio_cell_sample - One of the two example single-cell count matrices to be used for training 'scGPS' model
day_5_cardio_cell_sample - One of the two example single-cell count matrices to be used for 'scGPS' prediction
training_gene_sample - Input gene list for training 'scGPS', e.g. differentially expressed genes

On BioConductor:scGPS-1.27.0(bioc 3.24)scGPS-1.26.0(bioc 3.23)

singlecell clustering dataimport sequencing coverage openblas cpp

5.30 score 5 stars 9 scripts 2 mentions 38 exports 103 dependencies

Last updated from:5a0ba746f6. Checks:1 WARNING, 11 NOTE, 2 OK. Indexed: yes.

Target	Result	Time
bioc-checks	WARNING	360
linux-devel-arm64	NOTE	532
linux-devel-x86_64	NOTE	1209
source / vignettes	OK	1650
linux-release-arm64	NOTE	583
linux-release-x86_64	NOTE	669
macos-release-arm64	NOTE	449
macos-release-x86_64	NOTE	779
macos-oldrel-arm64	NOTE	335
macos-oldrel-x86_64	NOTE	999
windows-devel	NOTE	620
windows-release	NOTE	542
windows-oldrel	NOTE	558
wasm-release	OK	274

Exports:annotate_clusters bootstrap_parallel bootstrap_prediction calcDist calcDistArma clustering clustering_bagging CORE_bagging CORE_clustering CORE_subcluster distvec find_markers find_optimal_stability find_stability mean_cpp new_scGPS_object new_summarized_scGPS_object PCA plot_CORE plot_optimal_CORE plot_reduced predicting PrinComp_cpp rand_index rcpp_Eucl_distance_NotPar rcpp_parallel_distance reformat_LASSO sub_clustering subset_cpp summary_accuracy summary_deviance summary_prediction_lasso summary_prediction_lda top_var tp_cpp training tSNE var_cpp

Dependencies:abind BH Biobase BiocGenerics BiocParallel caret class cli clock codetools cpp11 data.table DelayedArray DESeq2 diagram digest dplyr dynamicTreeCut e1071 farver fastcluster foreach formatR futile.logger futile.options future future.apply generics GenomicRanges ggplot2 glmnet globals glue gower gtable hardhat ipred IRanges isoband iterators KernSmooth labeling lambda.r lattice lava lifecycle listenv locfit lubridate magrittr MASS Matrix MatrixGenerics matrixStats ModelMetrics nlme nnet numDeriv parallelly pillar pkgconfig plyr pROC prodlim progressr proxy purrr R6 RColorBrewer Rcpp RcppArmadillo RcppEigen RcppParallel recipes reshape2 rlang rpart S4Arrays S4Vectors S7 scales Seqinfo shape SingleCellExperiment snow SparseArray sparsevctrs SQUAREM stringi stringr SummarizedExperiment survival tibble tidyr tidyselect timechange timeDate tzdb utf8 vctrs viridisLite withr XVector

scGPS introduction

1. Installation instruction | 2. A simple workflow of the scGPS: | 2.1 Create scGPS objects | 2.2 Run prediction | 2.3 Summarise results | 3. A complete workflow of the scGPS: | 3.1 Identify clusters in a dataset using CORE | 3.2 Identify clusters in a dataset using SCORE (Stable Clustering at Optimal REsolution) | 3.3 Visualise all cluster results in all iterations | 3.4 Compare clustering results with other dimensional reduction methods (e.g., tSNE) | 3.5 Find gene markers and annotate clusters | 4. Relationship between clusters within one sample or between two samples | 4.1 Start the scGPS prediction to find relationship between clusters | 4.2 Display summary results for the prediction | 4.3 Plot the relationship between clusters in one sample | 4.3 Plot the relationship between clusters in two samples

Rendered fromvignette.Rmdusingknitr::rmarkdown

Last update: 2020-12-02
Started: 2017-11-30

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
add_import	add_import
annotate_clusters functionally annotates the identified clusters	annotate_clusters
BootStrap runs for both scGPS training and prediction with parallel option	bootstrap_parallel
BootStrap runs for both scGPS training and prediction	bootstrap_prediction
Compute Euclidean distance matrix by rows	calcDist
Compute Euclidean distance matrix by rows	calcDistArma
HC clustering for a number of resolutions	clustering
HC clustering for a number of resolutions	clustering_bagging
Main clustering SCORE (CORE V2.0) Stable Clustering at Optimal REsolution with bagging and bootstrapping	CORE_bagging
Main clustering CORE V2.0 updated	CORE_clustering
sub_clustering (optional) after running CORE 'test'	CORE_subcluster
One of the two example single-cell count matrices to be used for training 'scGPS' model	day_2_cardio_cell_sample
One of the two example single-cell count matrices to be used for 'scGPS' prediction	day_5_cardio_cell_sample
Compute Distance between two vectors	distvec
find marker genes	find_markers
Find the optimal cluster	find_optimal_stability
Calculate stability index	find_stability
Calculate mean	mean_cpp
new_scGPS_object	new_scGPS_object
new_summarized_scGPS_object	new_summarized_scGPS_object
PCA	PCA
Plot dendrogram tree for CORE result	plot_CORE
plot one single tree with the optimal clustering result	plot_optimal_CORE
plot reduced data	plot_reduced
Main prediction function applying the optimal ElasticNet and LDA models	predicting
Principal component analysis	PrinComp_cpp
Calculate rand index	rand_index
Function to calculate Eucledean distance matrix without parallelisation	rcpp_Eucl_distance_NotPar
distance matrix using C++	rcpp_parallel_distance
summarise bootstrap runs for Lasso model, from 'n' bootstraps	reformat_LASSO
sub_clustering for selected cells	sub_clustering
Subset a matrix	subset_cpp
get percent accuracy for Lasso model, from 'n' bootstraps	summary_accuracy
get percent deviance explained for Lasso model, from 'n' bootstraps	summary_deviance
get percent deviance explained for Lasso model, from 'n' bootstraps	summary_prediction_lasso
get percent deviance explained for LDA model, from 'n' bootstraps	summary_prediction_lda
select top variable genes	top_var
Transpose a matrix	tp_cpp
Main model training function for finding the best model that characterises a subpopulation	training
Input gene list for training 'scGPS', e.g. differentially expressed genes	training_gene_sample
tSNE	tSNE
Calculate variance	var_cpp