This software package is meant for comparing NGS short read mappers.
Artificial reads can be generated from genome sequences such that their true
mapping positions are known (scripts/uniform). Artificial point mutations
(scripts/multiple_mutate) and indels (scripts/indel) can be introduced.
By comparing the original read locations to the positions the reads were
assigned to by the mapper, the sensitivity and specificity of the mapper can
be assessed.
By running the mapper with different combinations of parameters, an optimal
parameter set can be found (scripts/cross_tab and scripts/table2calls).
TAPAS is compatible with all short read mappers by through a simple text
file which contains the mapping command and the relevant parameters. Instead
of the parameter values, bash variables (${variablename}) are used. They
are replaced with the mapping parameter values to be tested by TAPAS.
Additionally, reads can be mutated in a pattern similar to typical damage found
in ancient DNA (aDNA) (scripts/mapdamage2geomparam). The effect of this
mutations can be inspected as well.
All the scripts are designed with the UNIX philosophy "do one thing and do it well" in mind. Therefore an analysis consists of a pipeline of these scripts.
The universal data format is the text table, a simple text file containing data in whitespace- or tab-separated columns. The first line serves as header line with column names.
The availabe tools are described in-depth in the TAPAS manual.
The manual is located in the docs/ folder. Open the file docs/index.html
inside the TAPAS folder with your browser to view it.
The manual can also be found online, click here.
There is also one example script which summarises the steps of the manual.
It can be found at manual/example-manual.sh.
To get help about one specific TAPAS tool, execute the tool with the option
--help. Each tool prints then an extensive help page which explains how
to use it. Example: TAPAS/scripts/uniform --help, where TAPAS is the
folder you installed TAPAS to, prints the help of the artificial read
generation tool uniform.
This software package needs the following tools to be installed:
- R >= 3.2
- Python >= 3.4 and Pip, the Python package manager
- Standard GNU tools like
head,awk,sort,join,sed. These are included in all GNU/Linux and Mac distributions. On Windows you can install Cygwin to get these programs. However, this software package is not yet tested on Windows.
These are the commands needed to install these dependencies on your system. Select those which match your GNU/Linux distribution:
Ubuntu and derivatives, like Scientific Linux and Linux Mint:
sudo apt-get install r python3 python3-pip
CentOS or Red Hat Linux:
sudo yum --enablerepo=extras install epel-release
sudo yum install r python3 python3-pip
To install TAPAS, first download it from the Releases section
Then, execute the script
TAPAS/scripts/gen/install_dependencies
where you replace TAPAS by the folder where you installed TAPAS into. The
tool downloads and installs the needed R and Python packages. The packages
are installed inside the TAPAS folder and do no affect the rest of the system.
If you want to run TAPAS using R and Python packages which are installed
globally on your system, instead of the install_dependencies script, run
TAPAS/script/gen/gen-launchers.sh --ext-libs
To check if all dependencies are met, run
TAPAS/scripts/setup_check_dependencies
on your computer.
All scripts lie in the scripts/ folder. Read the manual to get an overview.
Additionally, each script can be executed with the --help switch to get
detailed information about the script's functionality and how it can be tailored
to your specific needs.