Skip to content

Latest commit

 

History

History
163 lines (126 loc) · 5.81 KB

parameter_estimation_increased_throughput.rst

File metadata and controls

163 lines (126 loc) · 5.81 KB
Raw

Simple Parameter Estimation

This is an example of how to configure a simple parameter estimation using pycotools. We first create a toy model for demonstration, then simulate some experimental data from it and fit it back to the model, using pycotools for configuration.

Configuring a model for parameter estimation

import os, glob
import pandas, numpy
import matplotlib.pyplot as plt
import seaborn
from pycotools3 import model, tasks, viz
seaborn.set_context(context='talk')     # set seaborn context for formatting output of plots

## Choose a directory for our model and analysis. Note this can be anywhere. 
working_directory = os.path.abspath('')

## In this model, A gets reversibly converted to B but the backwards reaction is additionally regulated by C.
## B is reversibly converted into C.
antimony_string = """
model simple_parameter_estimation()
    compartment Cell = 1;

    A in Cell;
    B in Cell;
    C in Cell;

    // reactions
    R1: A => B ; Cell * k1 * A;
    R2: B => A ; Cell * k2 * B * C;
    R3: B => C ; Cell * k3 * B;
    R4: C => B ; Cell * k4 * C;

    // initial concentrations
    A = 100;
    B = 1;
    C = 1;

    // reaction parameters
    k1 = 0.1;
    k2 = 0.1;
    k3 = 0.1;
    k4 = 0.1;
end
"""

# Create a path to a copasi file
copasi_file = os.path.join(working_directory, 'example_model.cps')

## build model
mod = model.loada(antimony_string, copasi_file)
assert isinstance(mod, model.Model)

## simulate some data, returns a pandas.DataFrame
data = mod.simulate(0, 20, 1)

## write data to file
experiment_filename = os.path.join(working_directory, 'experiment_data.txt')
data.to_csv(experiment_filename)

## We now have a model and some experimental data and can
## configure a parameter estimation

Serial Execution

As shown in other examples, you can run a single parameter estimation using PyCoTools as a controller by setting run_mode=True.

with tasks.ParameterEstimation.Context(mod, experiment_filename, context='s', parameters='g') as context:
    context.set('separator', ',')
    context.set('run_mode', True)
    context.set('randomize_start_values', True)
    context.set('method', 'genetic_algorithm')
    context.set('population_size', 100)
    context.set('lower_bound', 1e-1)
    context.set('upper_bound', 1e1)
    config = context.get_config()
pe = tasks.ParameterEstimation(config)

Parallel Execution

PyCoTools also enables you to increase parameter estimation throughput by automating the parallel execution of multiple model copies. To do this, we simply set run_mode='parallel'. With 'parallel' mode, there are a few additional settings that you need to know about.

  • copy_number is the number of model copies that you want to run
  • pe_number is the number of parameter estimations each of copy_number models will run in serial
  • nproc is the number of processes to use for getting through all copy_number models.

The configuration below will run 20 parameter estimations using a pool of 6 processes.

with tasks.ParameterEstimation.Context(mod, experiment_filename, context='s', parameters='g') as context:
    context.set('separator', ',')
    context.set('run_mode', 'parallel')
    context.set('copy_number', 20)
    context.set('pe_number', 1)
    context.set('max_active', 6)
    context.set('randomize_start_values', True)
    context.set('method', 'genetic_algorithm')
    context.set('population_size', 100)
    context.set('lower_bound', 1e-1)
    context.set('upper_bound', 1e1)
    config = context.get_config()
pe = tasks.ParameterEstimation(config)

Warning

:

If you set max_active to a number larger than the number of cores in your machine, although you will be running more models at once, you will be running each model more slowly. Running at full capacity, a single COPASI instance takes about 12% CPU. If you set max_active to 12 on an 8 core machine, each instance will run at about 4% CPU.

On a Computer Cluster

If you have access to a computer cluster, then PyCoTools already supports Slurm and SunGridEngine sheduling systems. If you are using slurm, set run_mode='slurm' and PyCoTools will submit copy_number jobs using sbatch. If you are using SunGridEngine then set run_mode='sge'.

with tasks.ParameterEstimation.Context(mod, experiment_filename, context='s', parameters='g') as context:
    context.set('separator', ',')
    context.set('run_mode', 'slurm') # or sge
    context.set('copy_number', 300)
    context.set('pe_number', 1)
    context.set('max_active', 6)
    context.set('randomize_start_values', True)
    context.set('method', 'genetic_algorithm')
    context.set('population_size', 100)
    context.set('lower_bound', 1e-1)
    context.set('upper_bound', 1e1)
    config = context.get_config()
pe = tasks.ParameterEstimation(config)

Note

PyCoTools will expect Copasi to already be available in the environment so that the command CopasiSE will produce Copasi's help message and not an error.

If you are using a scheduling system different to SGE or Slurm then you'll have to write your own wrapper, analogous to :pypycotools3.tasks.Run.submit_copasi_job_SGE and :pypycotools3.tasks.Run.submit_copasi_job_slurm. This shouldn't be too difficult. If you run into trouble please submit an issue on GitHub.