---
title: 'GeneNetwork: framework for web-based genetics'
tags:
  - bioinformatics
  - genetics
  - genomics
authors:
  - name: Zachary Sloan
    orcid: 0000-0002-8099-1363
    affiliation: University of Tennessee Health Science Center, USA
  - name: Danny Arends
    orcid: 0000-0001-8738-0162
    affiliation: Humboldt University, Berlin, Germany
  - name: Karl W. Broman
    orcid: 0000-0002-4914-6671
    affiliation: University of Wisconsin, USA
  - name: Arthur Centeno
    orcid: 0000-0003-3142-2081
    affiliation: University of Tennessee Health Science Center, USA
  - name: Nicholas Furlotte
    orcid: 0000-0002-9096-6276
  - name: Harm Nijveen
    orcid: 0000-0002-9167-4945
    affiliation: Wageningen University, The Netherlands
  - name: Lei Yan
    orcid: 0000-0001-5259-3379
    affiliation: University of Tennessee Health Science Center, USA
  - name: Xiang Zhou
    orcid: 0000-0002-4331-7599
    affiliation: University of Michigan
  - name: Robert W. Williams
    orcid: 0000-0001-8924-4447
    affiliation: University of Tennessee Health Science Center, USA
  - name: Pjotr Prins
    orcid: 0000-0002-8021-9162
    affiliation: University Medical Center Utrecht, The Netherlands, University of Tennessee Health Science Center, USA
date: 29 May 2016
bibliography: paper.bib
---

# Summary

GeneNetwork (GN) is a free and open source (FOSS) framework for
web-based genetics that can be deployed anywhere. GN allows biologists
to upload high-throughput experimental data, such as expression data
from microarrays and RNA-seq, and also `classic' phenotypes, such as
disease phenotypes.  These phenotypes can be mapped interactively
against genotypes using embedded tools, such as R/QTL [@Arends:2010]
mapping, interval mapping for model organisms and pylmm; an
implementation of FaST-LMM [@Lippert:2011] which is more suitable for
human populations and outbred crosses, such as the mouse diversity
outcross. Interactive D3 graphics are included from R/qtlcharts and
presentation-ready figures can be generated. Recently we have added
functionality for phenotype correlation [@Wang:2016] and network
analysis [@WGCNA:2008].

-![Mouse LMM mapping example](qtl2.png)

GN is written in python and javascript and contains a rich set of
tools and libraries that can be written in any computer language. A
full list of included software can be found in the package named
`genenetwork2' and defined in
[guix-bioinformatics](https://github.com/genenetwork/guix-bioinformatics/blob/master/gn/packages/genenetwork.scm). To
make it easy to install GN locally in a byte reproducible way,
including all dependencies and a 2GB MySQL test database (the full
database is 160GB and growing), GN is packaged with
[GNU Guix](https://www.gnu.org/software/guix/), as described
[here](https://github.com/genenetwork/genenetwork2/blob/master/doc/README.org).
GNU Guix deployment makes it feasible to deploy and rebrand GN
anywhere.

# Future work

More mapping tools will be added, including support for Genome-wide
Efficient Mixed Model Association (GEMMA). The
[Biodiallance genome browser](http://www.biodalliance.org/) is being
added as a Google Summer of Code project with special tracks related
to QTL mapping and network analysis. Faster LMM solutions are being
worked on, including GPU support.

A REST interface is being added so that data can be uploaded to a
server, analysis run remotely on high performance hardware, and
results downloaded and used for further analysis. This feature will
allow biologist-programmers to use R and Python on their computer and
execute computations on GN enabled servers.

# References