Skip to content

parthnatekar/Kinetic-Parameter-Estimation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

66 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Global + Local search for Kinetic Parameter Estimation

 

This code implements a parameter estimation pipeline for an MAPK cascade

Two algorithms are implmented - A genetic algorithm and the Nelder-Mead Simplex. These can be used in conjunction for any generic parameter estimation problem.

 

Workflow

Parameter search using local Nelder-Mead simplex

Parameter search using Genetic Algorithm

Combined estimation using GA for global search and NMS for local search

Results using only Nelder-Mead simplex or only GA

Results using only Nelder-Mead (Better for local search) Results using only GA (Better for global search)

 

Results using both algorithms

While the Nelder-Mead simplex is a good local search heuristic, Genetic Algorithms are good for global search. It then makes sense to apply these in sucession, getting global estimates from the Genetic Algorithm and fine-tuning these with the Nelder-Mead simplex

Results using both Algorithms together Results using both Algorithms together

 

Performance of the three search heuristics

drawing

Error for different algorithms

 

As can be seen from the images and table, GA for global search followed by NM for local search gives most accurate parameter estimates.