In LTLA/DropletTestFiles: Test Files for Single-Cell Droplet Utilities

knitr::opts_chunk$set(error=FALSE, message=FALSE, warning=FALSE)
library(BiocStyle)

Motivation

The r Biocpkg("DropletTestFiles") package contains files for testing droplet-based utilities, such as those in the r Biocpkg("DropletUtils") package. These files are literally the raw output of pipelines like 10X Genomics' CellRanger software suite, and are usually not in an (immediately) analysis-ready state. After all, the idea is to provide some material to test the utilities to get to such a state!

Functions

This package doesn't do anything except pull down and serve up files, so there's not much to talk about here. There are two convenience functions to help obtain content from r Biocpkg("ExperimentHub"). The first is to list all available resources managed by r Biocpkg("DropletTestFiles"):

library(DropletTestFiles)
out <- listTestFiles()
out

The second is to actually obtain a resource. This is provided in the form of a (read-only!) path on which further operations can be applied.

getTestFile(out$rdatapath[1], prefix=FALSE)

Currently, all of the files come from 10X Genomics datasets. As such, we will see a lot of filtered/raw count matrices, molecule information files and HDF5 barcode matrices. We refer readers to the (relevant section)[https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/output/overview] of the 10X Genomics website for more details.

Example

Here, we obtain a path to a filtered HDF5 matrix and read it in with a r Biocpkg("DropletUtils") function. This produces a SingleCellExperiment object for use in various Bioconductor pipelines.

library(DropletUtils)
path <- getTestFile("tenx-3.1.0-5k_pbmc_protein_v3/1.0.0/filtered.h5", prefix=TRUE)
sce <- read10xCounts(path, type="HDF5")
sce

Session information {-}

sessionInfo()

LTLA/DropletTestFiles documentation built on Sept. 18, 2020, 5:20 p.m.