hms-dbmi/dseqr

single-cell and bulk RNA-seq analyses from counts → pathways → drug candidates.

Dseqr

End-to-End RNA-Seq Analysis

Dseqr is a web application that helps you run 10X single-cell and bulk RNA-seq analyses from fastq → pathways → drug candidates.

💡 Read the Docs and Open Dseqr →

Local setup

# install
install.packages('remotes')
remotes::install_github('hms-dbmi/dseqr')

# initialize and run new project
library(dseqr)
project_name <- 'example'

# directory to store application and project files in
data_dir <- './dseqr'

run_dseqr(project_name, data_dir)

To enable bulk fastq.gz import, first build a kallisto index for quantification. To do so run:

# default as used by run_dseqr
indices_dir <- file.path(data_dir, '.indices_dir')

rkal::build_kallisto_index(indices_dir)

scRNA-seq fastqs

dseqr can directly import cellranger formatted count matrices. If you are starting from fastq files, first install kb-python:

# install kallisto|bustools wrapper (required)
pip install kb-python

Then run pseudo-quantification:

# download pre-built index (mouse or human)
dseqr::download_kb_index(indices_dir, species = 'human')

# run pseudo-quantification
data_dir <- 'path/to/folder/with/fastqs'
dseqr::run_kb_scseq(indices_dir, data_dir, species = 'human')

# clean intermediate files produced by kb
dseqr::clean_kb_scseq(data_dir)

The resulting cellranger formatted count matrix files will be in the data_dir subdirectory bus_output/counts_unfiltered/cellranger.

Prefer docker?

# pull image
docker pull alexvpickering/dseqr

# run at http://0.0.0.0:3838/ and keep data on exit
docker run -v /full/path/to/data_dir:/srv/dseqr \
-p 3838:3838 \
alexvpickering/dseqr R -e 'library(dseqr); run_dseqr("example", "/srv/dseqr")'

Host it

To spin up your own AWS infrastructure to host dseqr, see dseqr.aws →