mineapy 0.0.4

mineapy, Minimum network enrichment analysis in Python

MiNEApy

MiNEApy is a Python package for performing Minimum Network Enrichment Analysis (MiNEA), with a focus on thermodynamic and enrichment analysis in biological systems.

Vikash Pandey and Vassily Hatzimanikatis. Investigating the deregulation of metabolic tasks via Minimum Network Enrichment Analysis (MiNEA) as applied to nonalcoholic fatty liver disease using mouse and human omics data. DOI: https://doi.org/10.1371/journal.pcbi.1006760

Requirements

You will need to have Git-LFS installed to properly download some binary files:

git clone https://github.com/vpandey-om/mineapy.git /path/to/mineapy
cd /path/to/mineapy
git lfs install
git lfs pull

This module was developed in Python 3.7, and it is recommended to run Python 3.7 to ensure compatibility with commercial solvers such as Gurobi and CPLEX.

MiNEApy depends on several Python packages including:

• COBRApy
• pyTFA
• optlang

The installation process should handle these dependencies for you. However, if you are using a dedicated solver, you may need to install it separately. Supported solvers include GLPK, CPLEX, and Gurobi.

Setup

You will need to install conda and create an environment using Python 3.7:

conda create -n py37 python=3.7
conda activate py37

Using pip

To install MiNEApy with pip, you can use the requirements.txt file:

pip install -r requirements.txt

Alternatively, you can install the specific version directly from PyPI:

pip install mineapy==0.0.4

Installing from Source

If you prefer to install from the source:

conda create -n py37 python=3.7
conda activate py37

# Inside the py37 environment, download from GitHub and install.
git clone https://github.com/vpandey-om/mineapy.git /path/to/mineapy
cd /path/to/mineapy
pip install -e .

Quick Start

Two tutorial files detail the typical usage of the MiNEApy package. They can be found in the tutorials directory:

mineapy
└── tutorials
    ├── e_coli_core.py
    └── e_coli_gem.py

Here’s a quick example:

import mineapy
from cobra.io import load_matlab_model
import pandas as pd
from mineapy.core.taskEnrich import TaskEnrichment
from mineapy.core.thermo_model import ThermoModel_WithoutInfo
from mineapy.core.rxnExp import ReactionExp

# Load the e_coli_core model
cobra_model = load_matlab_model('./models/e_coli_core.mat')
genes = [g.id for g in cobra_model.genes]

# MiNEA parameters
path_to_params = './input/task_enrichment_params.yaml'

# Add condition- or context-specific expression data
context_df = pd.read_csv('./input/context.txt', sep='\t')
condition_df = pd.read_csv('./input/condition.txt', sep='\t')

# Get genes regulated between different conditions
gene_reg = {'gene_id': condition_df['geneid'].to_list(), 'fold_change': condition_df['fold change'].to_list(), 'up_cutoff': 1.35, 'down_cutoff': float(1/2.5)}
reg_analysis = ReactionExp(cobra_model, gene_reg=gene_reg)

# Set cutoff, e.g., 15% top and bottom in ranking
gene_exp = {'gene_id': context_df['geneid'].to_list(), 'exp_val': context_df['exp_val'].to_list(), 'high_cutoff': 0.15, 'low_cutoff': 0.15}
exp_analysis = ReactionExp(cobra_model, gene_exp=gene_exp)
params_rxns = {'high_rxns': exp_analysis.high_rxns, 'low_rxns': exp_analysis.low_rxns}

# Apply enrichment algorithms
task_enrich = TaskEnrichment(cobra_model, path_to_params, params_rxns)
task_enrich.run()

Usage

First, create your COBRApy model. Make sure to define the additional values required by pyTFA, as mentioned in the "Models" page of the documentation.

If you already have a Matlab model with thermodynamic data, you might want to use pytfa.io.import_matlab_model. Otherwise, refer to the COBRApy documentation to add the required properties.

If you're using a specific solver, don't forget to configure COBRApy by setting the solver property of your model. See the COBRApy solvers documentation for more information.

License

The software in this repository is licensed under the Apache-2.0 License. Please see the LICENSE file for more details.