PretextGraph

Converts bedgraph formatted data and embeds inside a Pretext contact map.

Bioconda

Note: currently seems not able to install this tool via Bioconda. (will be fixed soon I hope...)

All commandline Pretext tools for Unix (Linux and Mac) are available on bioconda.

The full suite of Pretext tools can be installed with

> conda install pretext-suite

Or, just PretextGraph can be installed with

> conda install pretextgraph

Usage

PretextGraph reads bedgraph formatted data from stdin, e.g:

> zcat bedgraph.file.gz | PretextGraph -i input.pretext -n "graph name"
> PretextGraph -i input.pretext -n "graph name" < /path/to/bedgraph.file
> bigWigToBedGraph bigwig.file /dev/stdout | PretextGraph -i input.pretext -n "graph name"

Inject extensions in to an existing .pretext file

Usage:

PretextGraph -i /path/to/input/.pretext/file -n "name_of_extension" [-o /path/to/the/output/.pretext/file] < /path/to/extension/data/file

NOTE: while using the bedgraph as the input for extension file, the newline character must be \n, and the separator must be \t. There is an example repeat_density.bedgraph file. The file is as follows:

#1_usercol      2_usercol       3_usercol       N_density
chr1    0       10000   5107
chr1    10000   20000   3579
chr1    20000   30000   4850
chr1    30000   40000   2643
chr1    40000   50000   2309
chr1    50000   60000   3308
chr1    60000   70000   3605
chr1    70000   80000   3908

Important: only non-negative integer data is supported. All values in 4th column should be int.

NOTE: Special process different extensions

There are different extensions types, such as the coverage, repeat_density, gap...

Currently, there are 3 ways to transform the data into extension graphs:

std::unordered_map<std::string, int> data_type_dic{  // use this data_type 
    {"default", 0, },
    {"repeat_density", 1},  // as this is counted in every single bin, so we need to normalise this within the bin
    {"gap", 2},  //
};

• 0: default, just add the weighted value of every bin to the graph->values[index];
• 1: repeat_density, before add the value of every bin to the graph->values[index], the value is normlised by the bin_size as the repeat.bedgraph counts number of repeat bps in one single bin;
• 2: gap, if there is gap in the related pixel, then graph->values[index] is set to 1 (no matter how many gaps), if no gaps within the range related with the pixel, the value is 0.

Options

-i input Pretext file, required. Sequence names in the Pretext file must match sequence names in the bedgraph data; although relative sort order is unimportant.
-i graph name, required. A name for the graph.

-o output Pretext file, optional. If no output is specified the graph data will be appended to the input file.

Requirments, running

4 cpu cores
128M RAM

Viewing graphs

PretextView version 0.1.2 or later.

Third-Party acknowledgements

PretextGraph uses the following third-party libraries:

• libdeflate
  
• stb_sprintf.h

Installation

meson

Requires:

• clang >= 11.0.0
• meson >= 0.57.1

git submodule update --init --recursive
env CXX=clang meson setup --buildtype=release --unity on --prefix=<installation prefix> builddir
cd builddir
meson compile
meson test
meson install

cmake

git submodule update --init --recursive
cd libdeflate
make || exit 1
cd ..
cmake -DCMAKE_BUILD_TYPE=Release -B build_cmake || exit 1
cd build_cmake 
make