incawrapper 1.0.0

General Repository for Omics Data Handling tools

incawrapper

incawrapper is a Python package which wraps for the matlab application INCA. INCA is a tool for 13C metabolic flux analysis. The incawrapper package allows to import data, setup the model and run INCA all from within Python. The results can be exported back to Python for further analysis and simply saved as .csv files. Furthermore, it is possible to export results from INCA runs entirely done through the GUI to Python.

What can the incawrapper do for me?

• Provide a Python interface to use INCA 100% independent of the INCA GUI
• Provide a data structure that can be imported to INCA
• Provide methods for exporting results from INCA to Python
• Provide methods for plotting results from INCA in Python
• Provide methods for creating INCA models with data, which can then be used in the INCA GUI
• Run both Isotopically Non-Stationary (INS) and Isotopically Stationary (IS) 13C-MFA
• Estimate fluxes and confidence intervals through the following INCA algorithms: estimate, parameter continuation, and Monte Carlo sampling

What can the incawrapper NOT do for me?

• Integration of NMR data
• Simulation of experiments
• Optimization of experimental design

How to use it?

This is an extremely quick show case of the incawrapper for more extensive examples please see our documentation page.

First, load your data typically atom mapped reactions, tracers information, flux measurements and MS measurements into pandas dataframes.

import pandas as pd
tracers_data = pd.read_csv("tracers.csv", 
    converters={'atom_mdv':ast.literal_eval, 'atom_ids':ast.literal_eval} # a trick to read lists from csv
)
reactions_data = pd.read_csv("reactions.csv")
flux_data = pd.read_csv("flux_measurements.csv")
ms_data = pd.read_csv("ms_measurements.csv", 
   converters={'labelled_atom_ids': ast.literal_eval} # a trick to read lists from csv
)

Then create the inca script and specify the options and which INCA algorithms to execute.

import incawrapper
output_file = "name/of/results/file.mat"
script = incawrapper.create_inca_script_from_data(reactions_data, tracers_data, flux_data, ms_data, experiment_ids=["exp1"])
script.add_to_block("options", incawrapper.define_options(fit_starts=5,sim_na=False))
script.add_to_block("runner", incawrapper.define_runner(output_file, run_estimate=True, run_simulation=True, run_continuation=True))

Now you are ready to run the inca script.

from incawrapper import run_inca
inca_directory = "path/to/inca/installation"
run_inca(script, INCA_base_directory=inca_directory)

INCA will now run in the background and execute the specified algorithms and store the results in the output_file. This file can be open in the INCA GUI (using Open Flux Map) or imported into Python:

res = incawrapper.INCAResults(output_file)
res.fitdata.fitted_parameters.head()

	type	id	eqn	val	std	lb	ub	free	...
0	Net flux	R1	A -> B	10	1e-05	9.99998	10	0	...
1	Net flux	R2 net	B <-> D	6.08415	0.0680021	5.9477	6.2182	1	...
2	Exch flux	R2 exch	B <-> D	6.62023	0.330634	6.00107	7.35286	1	...
3	Net flux	R3	B -> C + E	1.95792	0.0340011	1.8909	2.02615	1	...
4	Net flux	R4	B + C -> D + E + E	1.95792	0.0340011	1.8909	2.02615	0	...

Installation

For now, in order to install the incawrapper package, clone this repository onto your machine. Once ready, find the path to the base folder of your incawrapper clone and pip install the package like this >>> cd /path/to/incawrapper/base/folder >>> pip install ".[matlab]" Once released, incawrapper will be pip-installable. In a terminal, write >>> pip install incawrapper[matlab]

Supported Matlab and INCA versions

Both Matlab and INCA requires licenses which makes it difficult to automate testing of verison compatibility. For that reason will we only ensure compatibility with one INCA and one Matlab version. Currently supporting: Matlab 2023a, and INCA v2.2.

Documentation and examples

Example use cases and a description of the API can be found in our documentation (not uploaded yet).

Contributing

We welcome all contributions. Please follow the guidelines below when contributing code.

Quick start

1. Fork the repository
2. Clone the repository
3. Make a new branch from develop (see git model below) with your feature or fix
4. Submit a pull request

Git model

Please use the GitFlow model

Documentation

Please use the Numpy docstrings format

Commit messages

Please use the following standard for commit messages

• fix: ... for all commits that deal with fixing an issue
• feat: ... for all commits that deal with adding a new feature
• tests:... for all commits that deal with unit testing
• build:... for all commits that deal with the CI infrastructure and deployment

Pull requests

Please use the following PR title and description standards:

• The PR title should be short and descriptive. Work in progress reviews should be titled as WIP:... and all other should follow the above for commit messages.
• The PR description should describe 1) new features, 2) fixes, and 3) other changes

PR acceptance rules

In order to accept a PR, the following must be satisfied:

1. All new functions and classes have corresponding unit tests
2. All new functions and classes are documented using the correct style
3. All unit tests, linting tests, and integration tests pass
4. All new code is reviewed and approved by a repository maintainer