AtEnrich is a simple command-line tool for calculating enrichment of gene lists or clusters of genes across a variety of curated gene lists from functional genomic experiments.

This software runs on Unix operating systems (tested on Linux (Ubuntu) and Mac OS). It requires python 2.7 with the numpy, scipy, and pandas packages. These will be installed automatically through pip installation if not already installed.

The simplest way to install AtEnrich is from PyPI using pip:

pip install atenrich

Alternatively, install by running (from the top directory):

python setup.py install

AtEnrich can be run from the command line either for enrichment of a single gene list against a background gene list, or enrichment across a set of clusters. In list mode:

atenrich --mode list input_filename output_prefix

Here, the 'output_string' denotes the prefix used for files - the output files are then 'output_string_pvals.txt' and 'output_string_FE.txt'. 'list_filename' is a tab-delimited plaintext file of the form:

background_locus_id_0 list_of_interest_locus_id_0
background_locus_id_1 list_of_interest_locus_id_1
background_locus_id_2 list_of_interest_locus_id_2
background_locus_id_3 list_of_interest_locus_id_3
...                   ...

Locus identifiers are given by the Arabidopsis Genome Initiative (AGI) names (e.g. 'AT1G09570' for PHYA). Tabs separate the background gene list (left-hand column) and the list of interest (right-hand column). The genes of interest must be a subset of the background gene list.

In cluster mode, enrichment across all clusters can be evaluated:

atenrich --mode cluster input_filename output_prefix

Here, 'cluster_filename' is a tab-delimited plaintext file of the form:

locus_id_0  cluster_label_for_locus_id_0
locus_id_1  cluster_label_for_locus_id_1
locus_id_2  cluster_label_for_locus_id_2
locus_id_3  cluster_label_for_locus_id_3
...         ...

For the clustering of genes analysed in Seaton et al, 2017:

atenrich --mode cluster example_data/seaton2017_all_clusters_atenrich_input.csv example_data/cluster_mode_output

For just cluster 85 of this clustering:

atenrich --mode list example_data/seaton2017_cluster_85_atenrich_input.csv example_data/list_mode_output

The curated gene lists, which could not be supplied as part of the supplemental material, are supplied as part of the version of AtEnrich available to download from PyPI (https://pypi.org/project/atenrich/, in the atenrich/data/genelists folder).

• Daniel Seaton (https://github.com/danielseaton)

This software is licensed under the Apache 2.0 license (see LICENSE.txt).

Thanks to members of the Halliday Lab for assistance with testing.