Tools for Gene Ontology

Author	Haibao Tang (tanghaibao)
	DV Klopfenstein (dvklopfenstein)
	Brent Pedersen (brentp)
	Fidel Ramirez (fidelram)
	Aurelien Naldi (aurelien-naldi)
	Patrick Flick (patflick)
	Jeff Yunes (yunesj)
	Kenta Sato (bicycle1885)
	Chris Mungall (cmungall)
	Greg Stupp (stuppie)
	David DeTomaso (deto)
	Olga Botvinnik (olgabot)
Email	[email protected]
License	BSD

Description

This package contains a Python library to

• Process over- and under-representation of certain GO terms, based on Fisher's exact test. With numerous multiple correction routines including locally implemented routines for Bonferroni, Sidak, Holm, and false discovery rate. Also included are multiple test corrections from statsmodels: FDR Benjamini/Hochberg, FDR Benjamini/Yekutieli, Holm-Sidak, Simes-Hochberg, Hommel, FDR 2-stage Benjamini-Hochberg, FDR 2-stage Benjamini-Krieger-Yekutieli, FDR adaptive Gavrilov-Benjamini-Sarkar, Bonferroni, Sidak, and Holm.

• Process the obo-formatted file from Gene Ontology website. The data structure is a directed acyclic graph (DAG) that allows easy traversal from leaf to root.

• Read GO Association files:

  • Read GAF (Gene Association File) files.
  • Read NCBI's gene2go GO association file.

• Map GO terms (or protein products with multiple associations to GO terms) to GOslim terms (analog to the map2slim.pl script supplied by geneontology.org)

Installation

Make sure your Python version >= 2.7, install the latest stable version via PyPI:

easy_install goatools

To install the development version:

pip install git+git://github.com/tanghaibao/goatools.git

.obo file for the most current GO:

wget http://geneontology.org/ontology/go-basic.obo

.obo file for the most current GO Slim terms (e.g. generic GOslim) :

wget http://www.geneontology.org/ontology/subsets/goslim_generic.obo

Dependencies

• Simplest is to install via bioconda. See details here.

• To calculate the uncorrected p-values, there are currently twooptions:

  • fisher for calculating Fisher's exact test:

  easy_install fisher

  • fisher from SciPy's stats package

  • statsmodels (optional) for access to a variety of statistical tests for GOEA:

  easy_install statsmodels

• To plot the ontology lineage, install one of these two options:

  • Graphviz

    • Graphviz, for graph visualization.
    • pygraphviz, Python binding for communicating with Graphviz:

    easy_install pygraphviz

  • pydot, a Python interface to Graphviz's Dot language.

    • pyparsing is a prerequisite for pydot
    • Images can be viewed using either:
      • ImageMagick's display
      • Graphviz

Cookbook

run.sh contains example cases, which calls the utility scripts in the scripts folder.

Find GO enrichment of genes under study

See find_enrichment.py for usage. It takes as arguments files containing:

• gene names in a study
• gene names in population (or other study if --compare is specified)
• an association file that maps a gene name to a GO category.

Please look at tests/data/ folder to see examples on how to make these files. when ready, the command looks like:

python scripts/find_enrichment.py --pval=0.05 --indent data/study \
                                  data/population data/association

and can filter on the significance of (e)nrichment or (p)urification. it can report various multiple testing corrected p-values as well as the false discovery rate.

The "e" in the "Enrichment" column means "enriched" - the concentration of GO term in the study group is significantly higher than those in the population. The "p" stands for "purified" - significantly lower concentration of the GO term in the study group than in the population.

Important note: by default, find_enrichment.py propagates counts to all the parents of a GO term. As a result, users may find terms in the output that are not present in their association file. Use --no_propagate_counts to disable this behavior.

Read and plot GO lineage

See plot_go_term.py for usage. plot_go_term.py can plot the lineage of a certain GO term, by:

python scripts/plot_go_term.py --term=GO:0008135

This command will plot the following image.

Sometimes people like to stylize the graph themselves, use option --gml to generate a GML output which can then be used in an external graph editing software like Cytoscape. The following image is produced by importing the GML file into Cytoscape using yFile orthogonal layout and solid VizMapping. Note that the GML reader plugin may need to be downloaded and installed in the plugins folder of Cytoscape:

python scripts/plot_go_term.py --term=GO:0008135 --gml

Map GO terms to GOslim terms

See map_to_slim.py for usage. As arguments it takes the gene ontology files:

• the current gene ontology file go-basic.obo
• the GOslim file to be used (e.g. goslim_generic.obo or any other GOslim file)

The script either maps one GO term to its GOslim terms, or protein products with multiple associations to all its GOslim terms.

To determine the GOslim terms for a single GO term, you can use the following command:

python scripts/map_to_slim.py --term=GO:0008135 go-basic.obo goslim_generic.obo

To determine the GOslim terms for protein products with multiple associations:

python scripts/map_to_slim.py --association_file=data/association go-basic.obo goslim_generic.obo

Where the association file has the same format as used for find_enrichment.py.

The implemented algorithm is described in more detail at the go-perl documentation of map2slim.

Technical notes

Available statistical tests for calculating uncorrected p-values

There are currently two fisher tests available for calculating uncorrected p-values. Both fisher options from the fisher package and SciPy's stats package calculate the same pvalues, but provide the user an option in installing packages.

• fisher, fisher package's fisher.pvalue_population
• fisher_scipy_stats:SciPy stats package fisher_exact

Available multiple test corrections

We have implemented several significance tests:

• bonferroni, bonferroni correction
• sidak, sidak correction
• holm, hold correction
• fdr, false discovery rate (fdr) implementation using resampling

Additional methods are available if statsmodels is installed:

• sm_bonferroni, bonferroni one-step correction
• sm_sidak, sidak one-step correction
• sm_holm-sidak, holm-sidak step-down method using Sidak adjustments
• sm_holm, holm step-down method using Bonferroni adjustments
• simes-hochberg, simes-hochberg step-up method (independent)
• hommel, hommel closed method based on Simes tests (non-negative)
• fdr_bh, fdr correction with Benjamini/Hochberg (non-negative)
• fdr_by, fdr correction with Benjamini/Yekutieli (negative)
• fdr_tsbh, two stage fdr correction (non-negative)
• fdr_tsbky, two stage fdr correction (non-negative)
• fdr_gbs, fdr adaptive Gavrilov-Benjamini-Sarkar

In total 15 tests are available, which can be selected using option --method. Please note that the default FDR (fdr) uses a resampling strategy which may lead to slightly different q-values between runs.

iPython Notebooks

Run a Gene Ontology Enrichment Analysis (GOEA)

https://github.com/tanghaibao/goatools/blob/master/notebooks/goea_nbt3102.ipynb

Show many study genes are associated with RNA, translation, mitochondria, and ribosomal

https://github.com/tanghaibao/goatools/blob/master/notebooks/goea_nbt3102_group_results.ipynb

Report level and depth counts of a set of GO terms

https://github.com/tanghaibao/goatools/blob/master/notebooks/report_depth_level.ipynb

Find all human protein-coding genes associated with cell cycle

https://github.com/tanghaibao/goatools/blob/master/notebooks/cell_cycle.ipynb

Calculate annotation coverage of GO terms on various species

https://github.com/tanghaibao/goatools/blob/master/notebooks/annotation_coverage.ipynb

Determine the semantic similarities between GO terms

https://github.com/tanghaibao/goatools/blob/master/notebooks/semantic_similarity.ipynb

Want to Help?

If you add new code, please be sure to also add python tests which verify your code.

Items that we know we need include:

• Add code coverage runs

• Edit tests in the makefile under the comment, # TBD, suchthey run using nosetests

• Help setting up documentation. We are using Sphinx and Python docstrings to create documentation. For documentation practice, use make targets:

  make mkdocs_practice

  To remove practice documentation:

  make rmdocs_practice

  Once you are happy with the documentation do:

  make gh-pages

Reference

Haibao Tang et al. (2015). GOATOOLS: Tools for Gene Ontology. Zenodo. 10.5281/zenodo.31628.