A y-ome is a collection of genes in an organism that lack experimental evidence of function. This repository contains the software and analysis related to this publication on the y-ome of Escherichia coli:

Sankha Ghatak, Zachary A King, Anand Sastry, Bernhard O Palsson, The y-ome defines the 35% of Escherichia coli genes that lack experimental evidence of function, Nucleic Acids Research, Volume 47, Issue 5, 18 March 2019, Pages 2446–2454, https://doi.org/10.1093/nar/gkz030

Generating a y-ome

This repository includes a SQLite database (yome.db), scripts for populating the database (bin/), a Python package for querying the database (yome/), a set of Jupyter notebooks that contain the analyses described in the manuscript (notebooks), data files (data/), and a set of data source files (sources/).

Within each source folder in sources/, there is a load script and a set of data files. These load scripts are run sequentially (as defined in bin/load-db) to populate the database with source-specific annotations, then the script bin/load-yome generates the final y-ome gene list. The particular rules for identifying and assessing annotations from the data sources are described in comments within these files and also within the Methods section of the manuscript.

When the database loads successfully, bin/load-db updates the following data files in the data/ directory:

data/yome-genes.tsv - A tab-separated text file containing all genes from this analysis, their locus tags, primary name(s), and whether they are in the "y-ome", "well-annotated", or "excluded" category.

data/features.tsv.gz - A tab-separated text file containing all features extracted from the five knowledge bases. Each line contains a gene locus tag, the knowledge base name, the gene primary name(s), the feature type for that knowledge base, and the feature itself extracted as text.

data/no-information.tsv - A tab-separated text file containing a list of genes for which no information could be found in the five knowledge bases, with locus tags and EcoCyc primary names.

data/common-words.tsv - The most common words found in knowledge bases features for y-ome genes. The counts indicate the number of unique y-ome genes for which each phrase appears. Similar words are grouped into sets.

data/yome-transporters.tsv - A list of y-ome genes with indication in the knowledge bases that they might be transporters.

Run the workflow in Binder

Click the following link to launch the y-ome workflow in Binder. This will give you a fully interactive Jupyter Lab session:

View annotation and features for a y-ome gene

When you run the Binder server, you can run the notebook called "annotation report.ipynb" to quickly generate an overview of all the annotations for a given gene in the database. Just follow the instructions in the notebook.

Installation

The y-ome workflow includes a Python package that interacts with a SQLite database. If you want to use the packaged SQLite file (yome.db), be sure to install the yome package in develop mode:

pip install -e .

The bin/load-db script can be used to reload the database from the data in the sources directory.

EcoCyc data

EcoCyc data was downloaded using this tool:

https://github.com/zakandrewking/scrape_ecocyc