This repo contains the standardized analysis pipeline for 16S and metagenome data. It serves as a testing ground for what will be required to analyze around 500 samples.
- For suggestions, open questions and bugs please use the issue tracker.
- Different pipelines are managed via branches. The master branch points to the default pipeline.
- The project itself is managed in the Trello Board
For an examples analyzing a mock community see https://resendislab.github.io/microbiome.
If you want to perform Human sequence removal you will need bowtie2 installed and a human reference. The tools are bundled in the mbtools R package which also depends on all additional packages you need to run
the analyses. You will need to install bioconductor first followed by running
install.packages('devtools')
devtools::install_github('cdiener/microbiome/mbtools')in R.
Alternatively you can use the Docker image which is built and kept up to date automatically by watching this repo. For this you will require a local installation with Docker or a cloud instance which can run docker (for instance CoreOS VMs on AWS or Google Cloud).
First get the image with
docker pull cdiener/microbiomeYou can now start a Rstudio instance at port 8000 using
docker run -d -p 8000:8787 cdiener/microbiome
Now navigate your browser to :8000 (for instance localhost:8000 if you run docker on your machine) and you will be prompted for user credentials (use "rstudio" for user and password). All packages and dependencies are already installed here.
- How do I change the password? Login to the rstudio account with the default credentials (user and password are "rstudio"). Click on the "Tools" in the menu bar and choose "Shell". Type "passwd" and Enter. You will be promted for the old and new password.