BASiCS

Bayesian Analysis of Single-Cell Sequencing Data is an integrated Bayesian hierarchical model where:

• Cell-specific normalization constants are estimated as part of the model parameters.
• Technical variability is quantified based on spike-in genes that are artificially introduced to each analysed cells lysate.
• The total variability of the expression counts is decomposed into technical and biological components.

Installation

BASiCS will be submitted to BioConductor. In the meantime, the development version can be installed from GitHub:

# install.packages("devtools")
devtools::install_github("catavallejos/BASiCS", build_vignettes = TRUE)

This installation might fail if some of the dependency libraries are not yet installed. If so, please run the following lines and repeat the installation.

#library(devtools)
#source("http://bioconductor.org/biocLite.R")
#biocLite("BiocGenerics")
#biocLite("scran")
#install.packages("Rcpp")

After a successful installation, the BASiCS library can be loaded using[^footnoteInstall]

library(BASiCS)

[^footnoteInstall]: The warning "No methods found in "BiocGenerics"" might appear. Please ignore. BASiCS imports some of the generic functions provided by BiocGenerics that do not have any methods attached.

Troubleshooting

gfortran error

If you are unable to install BASiCS in a Mac OS X system, you might require an additional gfortran library.

• If you are running R versions 3.0.0 - R 3.3.0, this can be installed in the Mac terminal using the following command:

curl -O http://r.research.att.com/libs/gfortran-4.8.2-darwin13.tar.bz2
sudo tar fvxz gfortran-4.8.2-darwin13.tar.bz2 -C /

• If you are running R 3.4.0, you should install the official gfortran 6.1.0 binary for OS X El Capitan (10.11) from https://gcc.gnu.org/wiki/GFortranBinaries.

pandoc error

If you are not using RStudio, you might need to manually install pandoc in order to create the vignette provided for BASiCS

Additional installation issues/solutions can be found here: https://github.com/catavallejos/BASiCS/issues/9

How to use BASiCS?

• To detect highly and lowly variable genes within a populations of cells: please refer to the vignette

vignette('BASiCSIntro')

• To detect changes whose expression changes between 2 or more populations of cells (mean and over-dispersion), please refer to the supplementary material related to Vallejos et al., 2016

TODO: a quick start for BASiCS. Like vignette("some-stuff")

Development

Sanity check

In the folder of the development version from GitHub:

R CMD check .

Build from source

In the parent folder of the development version from GitHub:

R CMD build BASiCS

And run then

install.packages('BASiCS_x.x.x.tar.gz', repos=NULL)

on the generated file.

Author

Catalina A. Vallejos (cvallejos 'at' turing.ac.uk)

Collaborators

• Danilo Horta
• Nils Eling

References

• Vallejos et al. (2015). BASiCS: Bayesian Analysis of Single-Cell Sequencing Data
• Vallejos et al. (2016). Beyond comparisons of means: understanding changes in gene expression at the single cell level