GLnexus
Scalable datastore for population genome sequencing, with on-demand joint genotyping
|
From DNAnexus R&D: a scalable datastore for population genome sequencing, with on-demand joint genotyping. (GL, genotype likelihood)
This is an early-stage R&D project we're developing openly. The code doesn't yet do anything useful! There's a wiki project roadmap, which should be read in the spirit of "plans are worthless, but planning is insdispensable."

First install gcc 4.9, cmake
, libjemalloc-dev
, libboost-dev
, libzip-dev
, libsnappy-dev
, liblz4-dev
, libbz2-dev
. Then:
``` cmake -Dtest=ON . && make && ./unit_tests ```
Other dependencies (should be set up automatically by CMake):
Evolving developer documentation can be found on the project github page.
Status
, defined early in types.hS()
defined just below Status
Status
In order to do performance profiling on the code, you need to compile it with special c++ flags. This can be done as follows: ``` cmake -Dtest=ON -DCMAKE_BUILD_TYPE=Profile . ```
Running the glnexus_cli applet produces several outputs, one of which is a file called genotype.stacks
. It contains a list of common call stacks executed by the glnexus code, download it to your local machine. To manipulate this file, please install the FlameGraphs package. Here is an example for generating an SVG graph from the run results:
``` git clone https://github.com/brendangregg/FlameGraph grep glnexus genotype.stacks > X FlameGraph/flamegraph.pl –minwidth 0.5 X > genotype.svg ```
Documentation Last Updated:Wed Dec 2 19:05:25 UTC 2015