rfaster2

Get summary metrics about fastq data in R.

This package exports some of the functions of the Rust program faster2 in R. It uses Rust code (thanks to the extendr library) and is therefore pretty fast.

Install

If you don't have Rust, run this in your shell:

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

And then in R:

remotes::install_github('angelovangel/rfaster2')

If you don't want to setup Rust, use compiled GitHub releases. Binary packages are available for Linux (x86_64)

install.packages(
  "https://github.com/angelovangel/rfaster2/releases/download/v0.2.0/rfaster2_0.2.0_R_x86_64-pc-linux-gnu.tar.gz",
  dependencies = TRUE
)

and for macOS (aarch64, a.k.a. Apple Silicon)

install.packages(
  "https://github.com/angelovangel/rfaster2/releases/download/v0.2.0/rfaster2_0.2.0_macos_aarch64.tgz",
  dependencies = TRUE
)

The following additional libraries are needed:

dplyr
htmlwidgets
kableExtra
knitr
rmarkdown
scales
sparkline

Usage

All rfaster2 functions have the fq_ prefix. All of them take a path to a fastx file (fastq or fastq.gz) and return various metrics - reads, bases, N50, quality scores...

# get 'mean' qscores for the reads in a file
fq_quals(infile = path/to/fastqfile, phred = TRUE)

# get summary for a file (as a list object)
fq_summary(infile = path/to/fastqfile)

# get a data frame with statistics for a list of fastq files 
f <- list.files('path', pattern = 'fastq', full.names = T)
lapply(f, fq_summary) %>% dplyr::bind_rows()

An html report about all the fastq files in a directory can be generated using the fq_report function:

fq_report('path/to/fastq/files')

The html report will be saved in the R session working directory.