qtlsearch /

Filename Size Date modified Message
bin
examples
qtlsearch
35.1 KB
7.7 KB
10.5 KB
40 B
1.8 KB

QTLSearch

QTLSearch is piece of software to search for candidate causal genes in QTL studies by combining Gene Ontology annotations across many species, leveraging hierarchical orthologous groups.

First, a QTLSearch database is built (using qtlsearch-init) which contains OMA HOGs annotated with GO / ChEBI terms from the latest releases. This is based on a QTL and mapping file which the user provides.

Then, the search is performed using qtlsearch-run. A single database can be built for multiple QTL and then the search performed on these individually, enabling the use of a job scheduler. See examples from the paper to see how this was done using LSF.

Installation

Requires Python >= 3.6. Download the package from the PyPI, resolving the dependencies by using pip install qtlsearch.

Alternatively, clone this repository and install manually.

qtlsearch-init - Building a QTLSearch Database

Creates a database file qtlsearch.db in the current directory.

Usage

Required arguments: --species, --qtl, --annotation_map.

 usage: qtlsearch-init [-h] [--no_api_cache] [--api_cache_path API_CACHE_PATH]
                      [--api_endpoint API_ENDPOINT]
                      [--oma_version OMA_VERSION] [--data_path DATA_PATH]
                      --species SPECIES [SPECIES ...] --qtl QTL [QTL ...]
                      --annotation_map ANNOTATION_MAP [ANNOTATION_MAP ...]

Arguments

Quick reference table

Flag Default Description
--species Species to build database for.
--qtl List of paths to QTL files.
--annotation_map Mapping from trait in QTL file to annotation IDs (e.g., GO, ChEBI).
--data_path ./data Path to store downloaded data.
--api_cache_path ./api_cache Path to store OMA API cache.
--no_api_cache Boolean whether to cache API calls.
--oma_version Current Enables use of old releases, through database download.

Descriptions

--species

List of species to build the database for. Input as UniProt species codes (e.g., ARATH for Arabidopsis thaliana). These must be in the same order as the files in --qtl.

--qtl

List of paths to QTL file(s). These are tab-separated value files, format: <Trait, Chromosome, Start (Mbp), End (Mbp)>. An optional first column can be added containing QTL IDs, otherwise the ID will correspond to the line number in the results. Note: the files must not have header names; the file names must be in the same order as the species in --species.

--annotation_map

Mapping from trait in QTL file to annotation IDs (e.g., GO, ChEBI). These are tab-separated value files, format: <Trait, GO Term, ChEBI Term>. Note: the files must have header names "trait", "go", "chebi". The "chebi" column may be omitted if not required.

--data_path

Path to store downloaded data. This is safe to share between multiple runs of qtlsearch-init, unless the OMA database has been updated. Any updated species co-ordinates would then be incorrect.

--api_cache_path

Path to store OMA API cache. The expectation is to use this to share the data between multiple runs of qtlsearch-init.

--no_api_cache

Boolean whether to cache API calls. If set, this will not create the directory for persistent caching. If you don't expect to build multiple databases, it is best to disable this.

--oma_version

Enables use of old releases. Set to string of release, e.g., All.Sep2014 to use the species and HOGs from the September 2014 release of the OMA browser. This will download a very large database, instead of using the API.

Example Database Build

Here, a smaller database shall be built using the QTL, associated with Fructose or Galactose abundance, from the Lisec et al. dataset (in Arabidopsis thaliana [ARATH]) used in the paper.

Note: as we download (and cache) files from the UniProt-GOA and ChEBI databases, this will take quite some time. These files can be reused for further database builds, however.

Two files are required: one listing our trait to GO / ChEBI terms; the other listing the QTL.

QTL Listing

The following table shows the QTL listing, this is then stored as a tab seperated value formatted file, without headers, as qtl.tsv.

QTL ID Trait (Metabolite) Chromosome Start (Mbp) End (Mbp)
12 fructose 5 21.871688 24.169319
13 fructose 5 21.871688 23.594912
63 galactose 5 8.947509 11.245141
111 fructose 4 3.562645 7.833454
112 fructose 5 16.414812 17.563628

Trait Mapping

The following table shows the trait (metabolite) mapping used in the paper, this is then stored as a tab seperated value formatted file, with headers, as mapping.tsv.

trait go chebi
fructose GO:0046370 CHEBI:28757
galactose GO:0046369 CHEBI:28260

Build

The database can then be built as so:

qtlsearch-init --species ARATH --qtl qtl.tsv --annotation_map mapping.tsv

This will retrieve the current GO and ChEBI annotations which can be reused to build extra QTLSearch databases. The OMA API will be used to download the species protein co-ordinates required.

A qtlsearch.db file will be the output, placed in the current directory. This can then be used to search for your QTL in the next step.

qtlsearch-run -- Running QTLSearch

It is necessary to build a QTLSearch database prior to running the search. This is explained above.

Note: a single database can be built for multiple QTL and then the search performed on these individually, enabling the use of a job scheduler. See examples from the paper to see how this was done using LSF.

Usage

Required arguments: --species, --qtl, --db.

usage: qtlsearch-run [-h] [--data_path DATA_PATH] --species SPECIES
                 [SPECIES ...] --qtl QTL [QTL ...] --db DB [--with_p]
                 [--replicates REPLICATES] [--results RESULTS]

Arguments

Quick reference table

Flag Default Description
--db Path to database created with qtlsearch-init.
--species Species corresponding to QTL files.
--qtl List of paths to QTL files.
--results ./results.tsv Path to print the results table to.
--with_p Boolean whether to compute empirical p-values.
--replicates 1,000 Number of replicates for computing empirical distribution.
--data_path ./data Path where downloaded data was stored.

Descriptions

--db

Path to the database that is output from qtlsearch-init. If everything is completed in the same directory, without renaming the database file, this will be "qtlsearch.db".

--species

List of species to build the database for. Input as UniProt species codes (e.g., ARATH for Arabidopsis thaliana). These must be in the same order as the files in --qtl.

--qtl

List of paths to QTL file(s). These are tab-separated value files, format: <Trait, Chromosome, Start (Mbp), End (Mbp)>. An optional first column can be added containing QTL IDs, otherwise the ID will correspond to the line number in the results. Note: the files must not have header names; the file names must be in the same order as the species in --species.

--results

Path to print the results table to. Defaults as ./results.tsv. This is in TSV format and includes a header of column names in the first row.

--with_p

Boolean whether to compute the empirical p-values. This does take a long time, but the tool may still be useful without running this.

--replicates

Number of replicates for computing the empirical distribution, in order to estimate the p-values.

--data_path

Path to where the downloaded data was stored. This should be set to the same as it was for the call to qtlsearch-init.

Example Run

Following on from the example database build, we can now run to search for potential causal genes in the QTL associated with Fructose / Galactose abundance.

qtlsearch-run --species ARATH --qtl qtls.tsv --db qtlsearch.db

This will, by default, create the output in a file results.tsv in the current directory.

Change log

Version 1.0.3

Fix for pandas ambiguity error, caused by column / index having the same name.

Version 1.0.2

Includes a fix for larger HOGs.

Version 1.0.1

Initial public release.

License

QTLSearch is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

QTLSearch is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.

You should have received a copy of the GNU Lesser General Public License along with QTLSearch. If not, see http://www.gnu.org/licenses/.