eemisdada2

Packaging of dada2 for EEMiS, and other, users.

DADA2 is an error correction algorithm for Illumina amplicon reads. This repository offers four R scripts that lets you run the four main steps (cleaning and merging is performed by the same script) in the processing separately, saving intermediate results:

1. dada2filter: quality and length trimming
2. dada2errmodels: calculation of error models from the data
3. dada2cleanNmerge: sequence correction and merge
4. dada2bimeras: filter bimeras, create final tables

Requirements

DADA2 can be installed from the Bioconductor site.

DADA2 requires headers for the libcurl4-gnutls library. In Debian they can be installed with this command:

$ sudo apt install libcurl4-gnutls-dev

Installation

The scripts in this repository are easiest to install by cloning the repo and creating symlinks to the actual scripts in src/R.

Assuming you have $HOME/bin in your PATH variable and a dev directory in your home directory where you keep repositories like this, you would:

$ cd ~/dev/
$ git clone https://github.com/erikrikarddaniel/eemisdada2.git
$ cd ~/bin
$ ln -s ~/dev/eemisdada2/src/R/*
$ cd ~
$ dada2filter --version

(The last command just checks that everything is working.)

Now, you're ready to run the scripts, in the order specified above, from any directory. All scripts have a --help flag as documentation.

Automation via Make

In my biomakefiles repository, there's a makefile-based workflow implementation with documentation.