BioMASS - A Python Framework for Modeling and Analysis of Signaling Systems

Last update: Dec 27, 2022

Overview

BioMASS

Mathematical modeling is a powerful method for the analysis of complex biological systems. Although there are many researches devoted on producing models to describe dynamical cellular signaling systems, most of these models are limited and do not cover multiple pathways. Therefore, there is a challenge to combine these models to enable understanding at a larger scale. Nevertheless, larger network means that it gets more difficult to estimate parameters to reproduce dynamic experimental data needed for deeper understanding of a system.

To overcome this problem, we developed BioMASS, a Python framework for Modeling and Analysis of Signaling Systems. The BioMASS framework allows efficient optimization of multiple parameter sets simultaneously and generates the multiple parameter candidates that explain the signaling dynamics of interest. These parameter candidates can be further evaluated by their distribution and sensitivity analysis as a part of alternative information about the hidden regulatory mechanism of the system.

Features

Parameter estimation of ODE models
Local sensitivity analysis
Effective visualization of simulation results

Documentation

Online documentation is available at https://biomass-core.readthedocs.io/.

Installation

The BioMASS library is available at the Python Package Index (PyPI).

$ pip install biomass

BioMASS supports Python 3.7 or newer.

Also, we provide BioMASS docker images on DockerHub.

Example

Parameter estimation

from biomass import Model, optimize
from biomass.models import Nakakuki_Cell_2010

model = Model(Nakakuki_Cell_2010.__package__).create()

optimize(model, x_id=range(1, 11))

from biomass import run_simulation

run_simulation(model, viz_type="average", stdev=True)

Points (blue diamonds, EGF; red squares, HRG) denote experimental data, solid lines denote simulations.

Sensitivity analysis

from biomass import run_analysis

run_analysis(model, target="reaction", metric="integral", style="barplot")

Control coefficients for integrated pc-Fos are shown by bars (blue, EGF; red, HRG). Numbers above bars indicate the reaction indices, and error bars correspond to simulation standard deviation.

Citation

When using BioMASS, please cite the following paper:

Imoto, H., Zhang, S. & Okada, M. A Computational Framework for Prediction and Analysis of Cancer Signaling Dynamics from RNA Sequencing Data—Application to the ErbB Receptor Signaling Pathway. Cancers 12, 2878 (2020). https://doi.org/10.3390/cancers12102878

Author

Hiroaki Imoto

License

Apache License 2.0

Comments

Unable to run provided models

I am a beginner in both Python and data modeling and am having difficulties figuring out how to run this module based on the tutorials alone. I have been attempting to run the models provided but cannot get it to work. I had some help to install the module using pip, then I have downloaded the biomass_master file, and created a Python text file called biomass_test.py within the Nakakuki_Cell_2010 folder using Visual Studio Code. I have copied/pasted the commands `` from biomass import create_model from biomass.models import Nakakuki_Cell_2010

model = create_model(Nakakuki_Cell_2010.package, show_info=True) ``

within that text file, and ran it in a Python terminal in VS Code. It gives me the following error: FileNotFoundError: [Errno 2] No such file or directory: 'biomass\models\Nakakuki_Cell_2010\ode.py'`` The biomass_test.py and ode.py files are both in the Nakakuki_Cell_2010 folder. Could you please advise on what I may be doing wrong?
question

opened by lhdp0110 9
Unexpected behavior for reversible reaction
Description:

Text2Model of reversible reaction with one reactant and one product fails.

Reproduce:

Text file containing:

A + C <--> B A <--> B

Convert to model:

from biomass import Text2Model model = Text2Model("test.txt") model.convert(overwrite=True)

Raises:

DetectionError Unregistered words in line2: A <--> B
bug enhancement text2model
opened by formersbach 8
pytest on graphing fails with `pyvis==0.3`

Hi @formersbach, I noticed that pytest on graphing fails with the latest stable pyvis (version 0.3.0). When I restrict pyvis version via pyvis>=0.2.1,<0.3, we don't see any errors. Some updates in pyvis from v0.2 to v0.3 might affect the graph generation.
bug graph

opened by himoto 6
State transition
We have discussed the behavior of Text2Model in the issue #166. Accordingly, I added new reaction rule state_transition, in which simple mass action kinetics is applied when any rule words are not detected. In other words, this allows users to write a simple state transition, e.g., from A to B, as follows:

A <--> B

Also, this rule is direction-sensitive. If you use --> and <-->, the resulting reaction will be uni- and bi-directional, respectively.
text2model
opened by himoto 6
Using BioMASS without Graphviz
Discussed in https://github.com/biomass-dev/biomass/discussions/158

^{Originally posted by himoto July 23, 2022} BioMass requires users to manually install Graphviz for graph visualization but some packages derived from biomass, e.g., pasmpy, do not need it. Currently if we run import biomass without Graphviz, it will raise an ImportError. I would like to make this optional so that biomass-derived packages can be run without error even if Graphviz is not installed.

Ideas:

[x] Move pygraphviz and pyvis from requirements.txt to extras_require in setup.py

[x] Modify graph.py as follows:

def to_graph(self): try: import pygraphviz as pgv except ImportError: print("pygraphviz is required to run this function.") ... def dynamic_plot(self, ...): try: from pyvis.network import Network except ImportError: print("pyvis is required to run this function.")

@formersbach I would be grateful if you could give me comments on this. I have a branch working on this and if it's ok, I will send a PR.
graph
opened by himoto 6
Graph representation of biological model

Will add support for graphic representation of biological model. Requires graphviz program and pygraphviz package. Minor changes to exec_model.py to make create_graph callable on model object.

opened by formersbach 4
Error for the simulation example in the readme
The following error was produced when running the example in the readme

RuntimeWarning: invalid value encountered in double_scalars + (y[V.Fn] / x[C.KF31]) ** x[C.nF31]

The code:

from biomass import Model, optimize from biomass.models import Nakakuki_Cell_2010 model = Model(Nakakuki_Cell_2010.__package__).create() run_simulation(model, viz_type="average", stdev=True)

I think it is caused by an undefined parameter and thus, I propose changing the model used in the exampleto another model that does not require manual input of parameters, or to pre-input the default parameter for the model.

Please let me know if this is a misunderstanding.
opened by johannesnicolaus 3
Unexpected behavior in reaction_rules
Description:

Equation where 1 reactant reacts to 2 products is not covered by state_transition function but rather raises ValueError in _bind_and_dissociate. I was trying to create a transcription event with mass action kinetics. I'm not sure whether this is "right" or "wrong", maybe you have an idea @himoto !

Reproduce:

Text file containing:

nNfk --> nNfk + mIkb

Convert to model:

from biomass import Text2Model, create_model model = Text2Model("test.txt") model.convert(overwrite=True)

Observed behavior

Raises :

ValueError line1: nNfk <- Use a different name.

Desired behavior

ODE:

dmIkb/dt = kf1 * nNfk dnNfk/dt = 0
help wanted text2model
opened by formersbach 2
Type of `show_controls` in the example code
Hi @formersbach, I thinks the type of show_controls in the example code of dynamic_plot() should be boolean, not string:

model.dynamic_plot(save_dir='example_dir', file_name='nfkb_dynamic.html' show_controls='True', which_controls=['physics', 'layout'])

If yes, I have a fix-doc branch and fix this. Thank you!
documentation question
opened by himoto 2
Add note for `run_analysis` function

By default, run_analysis does not overwrite calculation results. If you would like to overwrite the results, you will need to set options['overwrite'] to False. This info should be emphasized at docstring.

opened by himoto 1
Phosphorylation generates duplicate ODE
Description:

If "is phosphorylated" reaction rule is used with a unphosphorylated species that has been previously used, a second ODE is mistakenly generated.

Reproduce:

Text file containing:

TNFR phosphorylates Ikk --> pIkk Ikk is phosphorylated <--> pIkk pIkk is phosphorylated <--> ppIkk

Convert to model:

from biomass import Text2Model, create_model model = Text2Model("test.txt") model.convert(overwrite=True) model = create_model(test)

Proposed changes:

Add missing format string in line 1099:

elif "dydt[V.{phosphorylated_form}]" in eq:

to

elif f"dydt[V.{phosphorylated_form}]" in eq:
bug text2model
opened by formersbach 1
Typo in state_transition rule
Description:

Modifier parsed with state_transition reaction rule leads to wrong kinetic information.

Reproduce:

Text file containing:

A + B --> A + pB

Convert to model:

from biomass import Text2Model, create_model model = Text2Model("test.txt") model.convert(overwrite=True) print(model.kinetics.rate)

Expected:

kf1 * pB * A

Observed:

kf1 * pB * A]

Proposed fix:

remove typo in line 1724 of reaction_rules.py: [f"{modifier}]" to [f"{modifier}"
opened by formersbach 2
Numerical structural identifiability analysis

I believe identifiability analysis is a very useful tool in model development and experiment design! Based on the work of Joubert and Stigter I have implemented the numerical part of their algorithm in a proof of concept script. Structural identifiability is a structural property of any model. To quote Joubert et al: "One way of characterizing structural identifiability is to say that at least 1 parameter has a confidence interval that spans the interval negative infinity to positive infinity. Any form of unidentifiability [...] calls into question the predictive power of a model and urges its user to interpret all results with caution" The idea of their algorithm is to analyze the so called parametric output sensitivity matrix (OSM). Full rank of the OSM is a sufficient condition for structural identifiability. Rank deficiency indicates non-influence by parameters or correlation between parameters. The rank of the OSM is calculated using singular value decomposition. Since all calculations are numerical and non-exhaustive there is a non-zero possibility of false negatives (identifiable parameter is deemed unidentifiable).

In principle no user input is required. A user defined observable matrix can be supplied however.

The script is ~ 100 lines long and requires the sympy package for symbolic calculations.

The only information required is a Text2Model object. Where in Pasmopy/Biomass do you think this should go? Should I just provide you the proof of concept script?
enhancement text2model

opened by formersbach 2

Releases(v0.10.0)

v0.10.0(Sep 22, 2022)
What's Changed

State transition by @himoto in https://github.com/biomass-dev/biomass/pull/169

Generalisation of mass action kinetics by @formersbach in https://github.com/biomass-dev/biomass/pull/172

Drop support for old template by @himoto in https://github.com/biomass-dev/biomass/pull/176

Fix typo in tutorial by @himoto in https://github.com/biomass-dev/biomass/pull/178

Fix module name: model_object by @himoto in https://github.com/biomass-dev/biomass/pull/181

Transfer graphing from BioMass to Text2Model by @formersbach in https://github.com/biomass-dev/biomass/pull/177

Update docs to be compatible with the newest features by @himoto in https://github.com/biomass-dev/biomass/pull/182

Fix f-string in is_phosphorylated rule by @himoto in https://github.com/biomass-dev/biomass/pull/184

Refactor code for sensitivity analysis by @himoto in https://github.com/biomass-dev/biomass/pull/186

Redirect rules by @himoto in https://github.com/biomass-dev/biomass/pull/188

Fix issue #189 by @formersbach in https://github.com/biomass-dev/biomass/pull/190

Fix docstring by @himoto in https://github.com/biomass-dev/biomass/pull/191

Fixes #192 by @formersbach in https://github.com/biomass-dev/biomass/pull/193

Update graphing tutorial to new graph function. by @formersbach in https://github.com/biomass-dev/biomass/pull/195

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.9.1...v0.10.0
Source code(tar.gz)
Source code(zip)
v0.9.1(Jul 27, 2022)
What's Changed

Enable running biomass without Graphviz by @himoto in https://github.com/biomass-dev/biomass/pull/162

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.9.0...v0.9.1
Source code(tar.gz)
Source code(zip)
v0.9.0(Jul 21, 2022)
What's Changed

Adding interactive plotting option. by @formersbach in https://github.com/biomass-dev/biomass/pull/144

Changed how graph is passed to pygraphvis, fixing issue #147 by @formersbach in https://github.com/biomass-dev/biomass/pull/148

Add graph visualization tutorial, resolves #145 by @formersbach in https://github.com/biomass-dev/biomass/pull/149

[pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in https://github.com/biomass-dev/biomass/pull/146

Fixes documentation and changes default save directory of graph.py by @formersbach in https://github.com/biomass-dev/biomass/pull/154

Remove exec by @himoto in https://github.com/biomass-dev/biomass/pull/157

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.8.0...v0.9.0
Source code(tar.gz)
Source code(zip)
v0.8.0(Jun 30, 2022)
What's Changed

Graph representation of biological model by @formersbach in https://github.com/biomass-dev/biomass/pull/131

Drop support of Python <= 3.7 by @himoto in https://github.com/biomass-dev/biomass/pull/132

Update graph.py by @himoto in https://github.com/biomass-dev/biomass/pull/133

From text to model by @himoto in https://github.com/biomass-dev/biomass/pull/136

Release v0.8.0 by @himoto in https://github.com/biomass-dev/biomass/pull/138

Fix error in multi_observables_options by @himoto in https://github.com/biomass-dev/biomass/pull/140

Fix dead link by @himoto in https://github.com/biomass-dev/biomass/pull/142

New Contributors

@formersbach made their first contribution in https://github.com/biomass-dev/biomass/pull/131

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.7.2...v0.8.0
Source code(tar.gz)
Source code(zip)
v0.7.2(Jun 16, 2022)
What's Changed

Make error messages more user friendly by @himoto in https://github.com/biomass-dev/biomass/pull/123

Fix #124 by @himoto in https://github.com/biomass-dev/biomass/pull/125

Release v0.7.2 by @himoto in https://github.com/biomass-dev/biomass/pull/128

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.7.1...v0.7.2
Source code(tar.gz)
Source code(zip)
v0.7.1(Jun 12, 2022)
What's Changed

Create tutorial by @himoto in https://github.com/biomass-dev/biomass/pull/118

Update logo by @himoto in https://github.com/biomass-dev/biomass/pull/119

Release v0.7.1 by @himoto in https://github.com/biomass-dev/biomass/pull/121

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.7.0...v0.7.1
Source code(tar.gz)
Source code(zip)
v0.7.0(Jun 1, 2022)
What's Changed

Add support for Python 3.10 by @himoto in https://github.com/biomass-dev/biomass/pull/110

Add CITATION.cff by @himoto in https://github.com/biomass-dev/biomass/pull/111

Add InitialPopulation by @himoto in https://github.com/biomass-dev/biomass/pull/113

Release v0.7.0 by @himoto in https://github.com/biomass-dev/biomass/pull/116

Renamed BioMASS modules:

| Old | New | | --- | ----- | | set_model.py | ode.py | | set_search_param.py | search_param.py | | fitness.py | problem.py |

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.6.2...v0.7.0
Source code(tar.gz)
Source code(zip)
v0.6.2(May 16, 2022)
What's Changed

Release v0.6.2 by @himoto in https://github.com/biomass-dev/biomass/pull/107

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.6.1...v0.6.2
Source code(tar.gz)
Source code(zip)
v0.6.1(Apr 26, 2022)
What's Changed

[pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in https://github.com/biomass-dev/biomass/pull/97

docs: git clone biomass by @himoto in https://github.com/biomass-dev/biomass/pull/98

Add model of the G1/S cell cycle transition by @himoto in https://github.com/biomass-dev/biomass/pull/99

Set 'LSODA' as default ODE solver by @himoto in https://github.com/biomass-dev/biomass/pull/100

Add model for the proliferation–quiescence decision by @himoto in https://github.com/biomass-dev/biomass/pull/101

Update example code for ExternalOptimizer by @himoto in https://github.com/biomass-dev/biomass/pull/103

Release v0.6.1 by @himoto in https://github.com/biomass-dev/biomass/pull/104

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.6.0...v0.6.1
Source code(tar.gz)
Source code(zip)
v0.6.0(Apr 1, 2022)
What's Changed

Release v0.6.0 by @himoto in https://github.com/biomass-dev/biomass/pull/95

[pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in https://github.com/biomass-dev/biomass/pull/87

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.5.5...v0.6.0
Source code(tar.gz)
Source code(zip)
v0.5.5(Dec 30, 2021)
What's Changed

Fix error message for incorrect visualization options by @himoto in https://github.com/biomass-dev/biomass/pull/72

Fix sensitivity analysis visualization problem by @himoto in https://github.com/biomass-dev/biomass/pull/81

Add more tests by @himoto in https://github.com/biomass-dev/biomass/pull/83

Release v0.5.5 by @himoto in https://github.com/biomass-dev/biomass/pull/84

[pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in https://github.com/biomass-dev/biomass/pull/70

Full Changelog: https://github.com/biomass-dev/biomass/compare/v0.5.4...v0.5.5
Source code(tar.gz)
Source code(zip)
v0.5.4(Oct 27, 2021)
Docs:

Add simulation results for each model, see here

Core functions:

Add method to check model indices

Solver:

Update function to equilibrate a system (#68)

Source code(tar.gz)
Source code(zip)
v0.5.3(Oct 11, 2021)
Docs:

Fix several docstrings

Update references

Requirements:

Set minimal numpy version to 1.17

Add typing_extensions

Other:

Add link to BioMASS docker images on DockerHub

Source code(tar.gz)
Source code(zip)

v0.5.2(Aug 20, 2021)

Template

Update viz.py

Use this template for creating models with biomass>=0.5.2

from matplotlib import pyplot as plt

from biomass.plotting import *

from .observable import Observable


class Visualization(Observable):
    """
    Plotting parameters for customizing figure properties.

    Attributes
    ----------
    cm : matplotlib.colors.ListedColormap (default: `plt.cm.get_cmap('tab10')`)
        Choosing colormaps for `cmap`.
    single_observable_options : list of SingleObservable
        Visualization options for time-course simulation (single-observable).
    multiple_observables_options : MultipleObservables
        Visualization options for time-course simulation (multi-observables).
    sensitivity_options : SensitivityOptions
        Visualization options for sensitivity analysis results.
    """

    def __init__(self):
        super().__init__()

        self.cm = plt.cm.get_cmap("tab10")
        self.single_observable_options = [
            SingleObservable(self.cm, obs_name) for obs_name in self.obs_names
        ]
        self.multiple_observables_options = MultipleObservables(self.cm)
        self.sensitivity_options = SensitivityOptions(self.cm)

    def get_single_observable_options(self):

        return self.single_observable_options

    def get_multiple_observables_options(self):

        return self.multiple_observables_options

    def get_sensitivity_options(self):

        return self.sensitivity_options

    @staticmethod
    def set_timecourse_rcParams():
        """figure/simulation"""
        plt.rcParams["font.size"] = 12
        plt.rcParams["axes.linewidth"] = 1.5
        plt.rcParams["xtick.major.width"] = 1.5
        plt.rcParams["ytick.major.width"] = 1.5
        plt.rcParams["lines.linewidth"] = 1.8
        plt.rcParams["lines.markersize"] = 12
        plt.rcParams["savefig.bbox"] = "tight"
        # plt.rcParams["savefig.format"] = "pdf"
        # plt.rcParams['font.family'] = 'Arial'
        # plt.rcParams['mathtext.fontset'] = 'custom'
        # plt.rcParams['mathtext.it'] = 'Arial:italic'

    @staticmethod
    def set_sensitivity_rcParams():
        """figure/sensitivity"""
        plt.rcParams["font.size"] = 12
        plt.rcParams["axes.linewidth"] = 1.2
        plt.rcParams["xtick.major.width"] = 1.2
        plt.rcParams["ytick.major.width"] = 1.2
        plt.rcParams["savefig.bbox"] = "tight"
        # plt.rcParams["savefig.format"] = "pdf"
        # plt.rcParams['font.family'] = 'Arial'

    @staticmethod
    def convert_species_name(name):
        """figure/sensitivity/initial_condition
        - Sensitivity for species with nonzero initial conditions
        """
        return name

Core functions
- Fix default workers in optimize function (#58)
Parameter estimation
- Fix a bug (#59)
Other
- Bump minimal sicpy version to 1.6
  - integrate.simps was renamed to integrate.simpson

Source code(tar.gz)
Source code(zip)

v0.5.1(Jul 14, 2021)
Template

Add new key: legend_kws

legend_loc becomes invalid

Several updates to default rcParams

Core functions

Remove save_format from arguments in run_simulation and run_analysis

To modify save_format, edit viz.py in the model.

Use x_id to specify the index (indices) of parameter sets when running optimize

start and end become invalid

Add create_metrics : Dict[str, Callable[[np.ndarray], Union[int, float]]], optional to run_analysis

Allowing users to crete signaling metrics for sensitivity analysis

Source code(tar.gz)
Source code(zip)

v0.5.0(Jun 23, 2021)

Dependencies
- Add pandas (See requirements.txt)
Template
- Update fitness.py
  - Create OptimizationProblem
  - Add bounds for external optimizer
- Update viz.py
  - Add figsize (default: (4, 3)) to timecourse_options and multiplot_options

biomass.result

Move visualization of estimated parameter sets to biomass.OptimizationResults.savefig()

>>> from biomass import Model, OptimizationResults
>>> from biomass.models import Nakakuki_Cell_2010
>>> model = Model(Nakakuki_Cell_2010.__package__).create()
>>> res = OptimizationResults(model)
>>> res.savefig(figsize=(16,5), boxplot_kws={"orient": "v"})

biomass.estimation

Create new class: ExternalOptimizer

>>> from scipy.optimize import differential_evolution
>>> from biomass import Model
>>> from biomass.models import Nakakuki_Cell_2010
>>> model = Model(Nakakuki_Cell_2010.__package__).create()
>>> optimizer = ExternalOptimizer(model, differential_evolution)
>>> res = optimizer.run(
...     model.problem.objective,
...     model.problem.bounds,
...     strategy="best2bin",
...     maxiter=100,
...     tol=1e-4,
...     mutation=0.1,
...     recombination=0.5,
...     disp=True,
...     polish=False,
...     workers=-1,
... )

differential_evolution step 1: f(x)= 7.05589 differential_evolution step 2: f(x)= 5.59166 differential_evolution step 3: f(x)= 2.80301 ... differential_evolution step 100: f(x)= 0.538524

>>> from biomass import run_simulation
>>> optimizer.import_solution(res.x, x_id=0)
>>> run_simulation(model, viz_type="0")

Source code(tar.gz)
Source code(zip)

v0.4.1(Jun 1, 2021)
Create docs: https://biomass-core.readthedocs.io/en/latest/

Core functions

Update default values: https://biomass-core.readthedocs.io/en/latest/modules/core.html

Optimization results

Set estimated params and initials to be plotted in one figure

Source code(tar.gz)
Source code(zip)

v0.4.0(May 10, 2021)

License
- BioMASS >= 0.4 is released under the Apache License 2.0.

Template

Don't use exec in name2idx

Use make_dataclass instead.

NAMES: List[str] = [...]

NUM: int = len(NAMES)

Parameters = make_dataclass(
    cls_name="Parameters",
    fields=[(name, int) for name in NAMES],
    namespace={"NAMES": NAMES, "NUM": NUM},
    frozen=True,
)

name2idx: Dict[str, int] = {k: v for v, k in enumerate(NAMES)}

C = Parameters(**name2idx)

del name2idx

Model construction

Use Model class

>>> from biomass import Model
>>> import your_model
>>> model = Model(your_model.__package__).create()

Sensitivity analysis
- Fix path to save sensitivity coefficients
  
  {your_model}/sensitivity_coefficients/{target}/{metric}.npy
Other
- Fix condition when exp.data to be plotted
- Update signaling_systems.py

Source code(tar.gz)
Source code(zip)

v0.3.5(Mar 31, 2021)

Models
- Add the insulin-dependent AKT pathway model
- Reference:
  - Kubota, H. et al. Temporal Coding of Insulin Action through Multiplexing of the AKT Pathway. Mol. Cell 46, 820–832 (2012). https://doi.org/10.1016/j.molcel.2012.04.018
- BioModels:
  - Kubota2012_InsulinAction_AKTpathway

Template for BioMASS model construction

Update reaction_network.py

from typing import Dict, List

from biomass.analysis.reaction import is_duplicate


class ReactionNetwork(object):
    """
    Reaction indices grouped according to biological processes.
    This is used for sensitivity analysis (target='reaction').
    """

    def __init__(self) -> None:
        self.reactions: Dict[str, List[int]] = {}

    def group(self):
        """
        Group reactions according to biological processes
        """
        biological_processes = []
        for process, indices in self.reactions.items():
            if not isinstance(indices, list):
                raise TypeError("Use list for reaction indices in {}".format(process))
            biological_processes.append(indices)

        if not is_duplicate(self.reactions, biological_processes):
            return biological_processes

Output

Update _write_best_fit_param.py

Example:

# parameter set: 9


def param_values():
    x = [0] * C.NUM
    x[C.V1] = 9.152e-02
    x[C.Km1] = 3.761e+02
    x[C.V2] = 2.200e-01
    x[C.Km2] = 3.500e+02
    x[C.V3] = 7.200e-01
    x[C.Km3] = 1.600e+02
    x[C.V4] = 6.480e-01
    x[C.Km4] = 6.000e+01
    x[C.V5] = 6.913e+01
    x[C.Km5] = 2.727e+01
    x[C.V6] = 6.913e+01
    x[C.Km6] = 2.727e+01
    x[C.KimERK] = 1.200e-02
    x[C.KexERK] = 1.800e-02
    x[C.KimpERK] = 1.200e-02
    x[C.KexpERK] = 1.800e-02
    x[C.KimppERK] = 1.100e-02
    x[C.KexppERK] = 1.300e-02
    x[C.V10] = 4.275e+02
    x[C.Km10] = 1.469e+01
    x[C.n10] = 2.593e+00
    x[C.p11] = 2.191e-04
    x[C.p12] = 1.156e-03
    x[C.p13] = 6.951e-04
    x[C.V14] = 3.090e-01
    x[C.Km14] = 8.111e+02
    x[C.V15] = 2.049e+02
    x[C.Km15] = 7.803e-03
    x[C.p16] = 2.570e-04
    x[C.p17] = 9.630e-05
    x[C.KimDUSP] = 1.832e-03
    x[C.KexDUSP] = 4.328e-02
    x[C.KimpDUSP] = 1.832e-03
    x[C.KexpDUSP] = 4.328e-02
    x[C.V20] = 3.512e-01
    x[C.Km20] = 1.321e+05
    x[C.V21] = 5.249e-02
    x[C.Km21] = 1.184e-03
    x[C.p22] = 2.570e-04
    x[C.p23] = 9.630e-05
    x[C.V24] = 3.440e+00
    x[C.Km24] = 1.912e+05
    x[C.V25] = 1.109e+02
    x[C.Km25] = 4.527e+00
    x[C.KimRSK] = 2.519e-03
    x[C.KexRSK] = 1.024e-02
    x[C.V27] = 1.390e+01
    x[C.Km27] = 4.619e+03
    x[C.V28] = 4.295e-01
    x[C.Km28] = 1.165e+00
    x[C.V29] = 9.103e-02
    x[C.Km29] = 6.333e+03
    x[C.V30] = 5.958e+01
    x[C.Km30] = 7.801e+00
    x[C.V31] = 5.691e+01
    x[C.Km31] = 1.543e-02
    x[C.n31] = 1.713e+00
    x[C.p32] = 1.100e-03
    x[C.p33] = 1.959e-03
    x[C.p34] = 7.572e-04
    x[C.V35] = 5.050e+00
    x[C.Km35] = 1.795e+03
    x[C.V36] = 1.218e-01
    x[C.Km36] = 2.394e+03
    x[C.V37] = 1.178e+03
    x[C.Km37] = 1.062e+02
    x[C.p38] = 2.570e-04
    x[C.p39] = 9.630e-05
    x[C.KimFOS] = 1.205e-01
    x[C.KexFOS] = 1.590e-01
    x[C.KimpcFOS] = 1.205e-01
    x[C.KexpcFOS] = 1.590e-01
    x[C.V42] = 9.649e-02
    x[C.Km42] = 5.639e+02
    x[C.V43] = 3.568e-03
    x[C.Km43] = 1.382e+02
    x[C.V44] = 1.290e-03
    x[C.Km44] = 2.725e-02
    x[C.p45] = 2.570e-04
    x[C.p46] = 9.630e-05
    x[C.p47] = 2.140e-01
    x[C.m47] = 1.288e+01
    x[C.p48] = 3.351e-02
    x[C.p49] = 4.279e-02
    x[C.m49] = 2.331e+00
    x[C.p50] = 3.491e-02
    x[C.p51] = 1.053e-03
    x[C.m51] = 4.536e+00
    x[C.p52] = 2.140e-01
    x[C.m52] = 1.288e+01
    x[C.p53] = 3.351e-02
    x[C.p54] = 4.279e-02
    x[C.m54] = 2.331e+00
    x[C.p55] = 3.491e-02
    x[C.p56] = 1.053e-03
    x[C.m56] = 4.536e+00
    x[C.V57] = 2.366e+02
    x[C.Km57] = 7.976e-04
    x[C.n57] = 1.649e+00
    x[C.p58] = 2.471e-09
    x[C.p59] = 2.889e-06
    x[C.p60] = 6.575e-02
    x[C.p61] = 2.517e+00
    x[C.KimF] = 2.058e-02
    x[C.KexF] = 7.998e-02
    x[C.p63] = 1.731e+00
    x[C.KF31] = 3.337e-03
    x[C.nF31] = 1.001e+00
    x[C.a] = 3.380e+02
    x[C.Vn] = 2.200e-01
    x[C.Vc] = 9.400e-01
    x[C.Ligand] = 0.000e+00
    x[C.EGF] = 0.000e+00
    x[C.HRG] = 1.000e+00
    x[C.no_ligand] = 2.000e+00


def initial_values():
    y0 = [0] * V.NUM
    y0[V.CREBn] = 1.000e+03
    y0[V.ERKc] = 9.600e+02
    y0[V.Elk1n] = 1.510e+03
    y0[V.RSKc] = 3.530e+02

Visualization
- Close #45
Other
- Update function to get the steady state for the untreated condition
  - Use scipy.integrate.ode
- Use math.isnan for single floats
- code refactoring
  - analysis/
  - estimation/

Source code(tar.gz)
Source code(zip)

v0.3.4(Mar 9, 2021)
Updates:

Parameter estimation

Fix search bounds, mutation and popsize in the differential evolution method.

Models

Move ODE solvers to dynamics/

Move search_param checkers to estimation/

Test

Update test paramter sets

Source code(tar.gz)
Source code(zip)

v0.3.3(Feb 16, 2021)

Update `optimize`

Add new option: initial_threshold
- initial_threshold : float (default: 1e12)
Threshold on objective function value used to generate initial population. Default value is 1e12 (numerically solvable).

def optimize(
    model: ModelObject,
    start: int,
    end: Optional[int] = None,
    options: Optional[dict] = None,
) -> None:
    """
    Run GA for parameter estimation.

    Paremters
    ---------
    model : ModelObject
        Model for parameter estimation.

    start : int
        Index of parameter set to estimate.

    end : int, optional
        When `end` is specified, parameter sets from `start` to `end` will be estimated.

    options : dict, optional
        popsize : int (default: 5)
            A multiplier for setting the total population size.
            The population has popsize * len(search_param) individuals.

        max_generation : int (default: 10000)
            Stop optimization if Generation > max_generation.

        initial_threshold : float (default: 1e12)
            Threshold on objective function value used to generate initial population.
            Default value is 1e12 (numerically solvable).

        allowable_error : float (default: 0.0)
            Stop optimization if Best Fitness <= allowable_error.

        local_search_method : str (default: 'mutation')
            Method used in local search. Should be one of
            * 'mutation' : NDM/MGG
            * 'Powell' : Modified Powell method
            * 'DE' : Differential Evolution (strategy: best2bin)

        n_children : int (default: 200)
            (method='mutation') The number of children generated in NDM/MGG.

        maxiter : int (default: 10)
            (method='Powell' or 'DE') The maximum number of iterations
            over which the entire population is evolved.

        workers : int (default: -1 if `end` is None else 1)
            (method='DE') The population is subdivided into workers sections and
            evaluated in parallel (uses multiprocessing.Pool). Supply -1 to use
            all available CPU cores. Set workers to 1 when searching multiple
            parameter sets simultaneously.

        overwrite : bool (default: False)
            If True, the out/n folder will be overwritten.

    Example
    -------
    >>> from biomass.models import Nakakuki_Cell_2010
    >>> from biomass import optimize
    >>> model = Nakakuki_Cell_2010.create()
    >>> optimize(
            model=model, start=1, end=10,
            options={
                'max_generation': 10000,
                'allowable_error': 0.5
            }
        )
    """
    ...

Update `run_analysis`

def run_analysis(
    model: ModelObject,
    target: str,
    metric: str = "integral",
    style: str = "barplot",
    save_format: str = "pdf",
    options: Optional[dict] = None,
) -> None:
    """
    Perform sensitivity analysis to identify critical parameters, species or
    reactions in the complex biological network.

    The sensitivity S(y,x) was calculated according to the following equation:
    S(y,x) = d ln(yi) / d ln (xj), where yi is the signaling metric and xj is
    each nonzero species, parameter value or reaction rate.

    Paremters
    ---------
    model : ModelObject
        Model for sensitivity analysis.

    target : str
        * 'reaction'
        * 'initial_condition'
        * 'parameter'

    metric : str (default: 'integral')
        * 'maximum' : The maximum value.
        * 'minimum' : The minimum value.
        * 'argmax' : The time to reach the maximum value.
        * 'argmin' : The time to reach the minimum value.
        * 'timepoint' : The simulated value at the time point set via options['timepoint'].
        * 'duration' :  The time it takes to decline below the threshold set via options['duration'].
        * 'integral' : The integral of concentration over the observation time.

    style : str (default: 'barplot')
        * 'barplot'
        * 'heatmap'

    save_format : str (default: "pdf")
        Either "png" or "pdf", indicating whether to save figures
        as png or pdf format.

    options : dict, optional
        show_indices : bool (default: True)
            (target == 'reaction') Set to True to put reaction index on each bar.

        excluded_params : list of strings
            (target == 'parameter') List of parameters which are not used for analysis.

        excluded_initials : list of strings
            (target == 'initial_condition') List of species which are not used for analysis.

        timepoint : int (default: model.sim.t[-1])
            (metric == 'timepoint') Which timepoint to use.

        duration : float (default: 0.5)
            (metric == 'duration') 0.1 for 10% of its maximum.

    Example
    -------
    >>> from biomass.models import Nakakuki_Cell_2010
    >>> from biomass import run_analysis
    >>> model = Nakakuki_Cell_2010.create()

    1. Parameter
    >>> run_analysis(
            model,
            target='parameter',
            options = {
                'excluded_params': [
                    'a', 'Vn', 'Vc', 'Ligand', 'EGF', 'HRG', 'no_ligand'
                ]
            }
        )

    2. Initial condition
    >>> run_analysis(
            model,
            target='initial_condition',
        )

    3. Reaction
    >>> run_analysis(
            model,
            target='reaction',
        )

    """
    ...

Source code(tar.gz)
Source code(zip)

v0.3.2(Jan 27, 2021)

Coding style

Code is formatted with black. Coding style is checked with flake8 and isort.

$ pip install pre-commit

$ pre-commit install

Display model information

Rename function: ~~show_properties~~ → show_info

from biomass.models import Nakakuki_Cell_2010

Nakakuki_Cell_2010.show_info()

↓

Nakakuki_Cell_2010 information
------------------------------
36 species
115 parameters, of which 75 to be estimated

Add new method to OptimizationResults: `trace_obj`

Visualize objective function traces for different optimization runs.

from biomass.models import Nakakuki_Cell_2010
from biomass.result import OptimizationResults

model = Nakakuki_Cell_2010.create()
res = OptimizationResults(model)

res.trace_obj()

Source code(tar.gz)
Source code(zip)

v0.3.1(Jan 15, 2021)
Updates:

New model

Dynamic pathway model of TNFα-induced NFκB signal transduction

Oppelt, A. et al. Model-based identification of TNFα-induced IKKβ-mediated and IκBα-mediated regulation of NFκB signal transduction as a tool to quantify the impact of drug-induced liver injury compounds. npj Syst. Biol. Appl. 4, 23 (2018). https://doi.org/10.1038/s41540-018-0058-z

Sensitivity analysis

New signaling metrics

'argmax' : The time to reach the maximum value.

'argmin' : The time to reach the minimum value.

'timepoint' : The simulated value at the time point set via options['timepoint'].

'duration' : The time it takes to decline below the threshold set via options['duration'].

Parameter estimation

Set mutation constant (differential weight: F) to 0.1

Set recombination constant (crossover probability: CR) to 0.9

Source code(tar.gz)
Source code(zip)

v0.3.0(Jan 6, 2021)

Updates:

Use BioMassModel for the first argument of core functions

Example usage

Create an executable model

from biomass.models import Nakakuki_Cell_2010

model = Nakakuki_Cell_2010.create()

Estimate model parameters from experimental data

from biomass import optimize

optimize(
    model=model, start=1, options={
        "popsize": 3,
        "max_generation": 1000,
        "allowable_error": 0.5,
        "local_search_method": "DE",
    }
)

Export optimized parameters in CSV format

from biomass.result import OptimizationResults

res = OptimizationResults(model)
res.to_csv()

Visualize simulation results

from biomass import run_simulation

run_simulation(model, viz_type='average', show_all=False, stdev=True)

Perform sensitivity analysis

from biomass import run_analysis

run_analysis(model, target='reaction', metric='integral', style='barplot')

Source code(tar.gz)
Source code(zip)

v0.2.3(Dec 26, 2020)
Small fixes

Update ODE solver

From scipy.integrate.ode to scipy.integrate.solve_ivp

Source code(tar.gz)
Source code(zip)

v0.2.2(Dec 14, 2020)

Updates:

Added new local search method: "Powell" and "DE"

def optimize(model, *args, options: Optional[dict] = None) -> None:
    """
    Run GA for parameter estimation.

    Paremters
    ---------
    model : module
        Model for parameter estimation.

    options: dict, optional
        popsize : int (default: 5)
            A multiplier for setting the total population size.
            The population has popsize * len(search_param) individuals.

        max_generation : int (default: 10000)
            Stop if Generation > max_generation.

        allowable_error : float (default: 0.0)
            Stop if Best Fitness <= allowable_error.

        local_search_method : str (default: 'mutation')
            Method used in local search. Should be one of
            - 'mutation' : NDM/MGG
            - 'Powell' : Modified Powell method
            - 'DE' : Differential Evolution (strategy: best2bin)

        n_children : int (default: 200)
            (method='mutation') The number of children generated in NDM/MGG.

        workers : int (default: -1 if len(args) == 1 else 1)
            (method='DE') The population is subdivided into workers sections and
            evaluated in parallel (uses multiprocessing.Pool). Supply -1 to use
            all available CPU cores. Set workers to 1 when searching multiple
            parameter sets simultaneously.

        overwrite : bool (default: False)
            If True, the out/n folder will be overwritten.
    """

Example usage

from biomass.models import Nakakuki_Cell_2010
from biomass import optimize

optimize(
    Nakakuki_Cell_2010, 1, options={
        'popsize': 3,
        'allowable_error': 0.5,
        'local_search_method': 'DE'
    }
)

Source code(tar.gz)
Source code(zip)

v0.2.1(Nov 25, 2020)
Fix:

result.py

Source code(tar.gz)
Source code(zip)
v0.2.0(Nov 21, 2020)
Updates:

Change sim.normalization type from bool to dict (nested):

def __init__(self): super().__init__(perturbation={}) self.normalization = {} for obs_name in observables: self.normalization[obs_name] = { 'timepoint' : None, 'condition' : [] }

'timepoint' : Optional[int]

The time point at which simulated values are normalized. If None, the maximum value will be used for normalization.

'condition' : list of strings

The experimental conditions to use for normalization. If empty, all conditions defined in sim.conditions will be used.

Move documentation to https://github.com/okadalabipr/biomass_docs.jl

Source code(tar.gz)
Source code(zip)
v0.1.2(Nov 12, 2020)

Source code(tar.gz)
Source code(zip)
v0.1.1(Oct 16, 2020)

Source code(tar.gz)
Source code(zip)

BioMASS - A Python Framework for Modeling and Analysis of Signaling Systems

Related tags

Overview

BioMASS

Features

Documentation

Installation

Example

Parameter estimation

Sensitivity analysis

Citation

Author

License

Comments

Description:

Reproduce:

Raises:

Discussed in https://github.com/biomass-dev/biomass/discussions/158

Description:

Reproduce:

Observed behavior

Desired behavior

Description:

Reproduce:

Proposed changes:

Description:

Reproduce:

Expected:

Observed:

Proposed fix:

Releases(v0.10.0)

v0.10.0(Sep 22, 2022)

What's Changed

v0.9.1(Jul 27, 2022)

What's Changed

v0.9.0(Jul 21, 2022)

What's Changed

v0.8.0(Jun 30, 2022)

What's Changed

New Contributors

v0.7.2(Jun 16, 2022)

What's Changed

v0.7.1(Jun 12, 2022)

What's Changed

v0.7.0(Jun 1, 2022)

What's Changed

v0.6.2(May 16, 2022)

What's Changed

v0.6.1(Apr 26, 2022)

What's Changed

v0.6.0(Apr 1, 2022)

What's Changed

v0.5.5(Dec 30, 2021)

What's Changed

v0.5.4(Oct 27, 2021)

v0.5.3(Oct 11, 2021)

v0.5.2(Aug 20, 2021)

v0.5.1(Jul 14, 2021)

v0.5.0(Jun 23, 2021)

v0.4.1(Jun 1, 2021)

v0.4.0(May 10, 2021)

v0.3.5(Mar 31, 2021)

v0.3.4(Mar 9, 2021)

Updates:

Parameter estimation

Models

Test

v0.3.3(Feb 16, 2021)

Update optimize

Update run_analysis

v0.3.2(Jan 27, 2021)

Coding style

Display model information

Add new method to OptimizationResults: trace_obj

v0.3.1(Jan 15, 2021)

Updates:

New model

Sensitivity analysis

New signaling metrics

Parameter estimation

Update `optimize`

Update `run_analysis`

Add new method to OptimizationResults: `trace_obj`