We, Prof. Steven Cramer group at Rensselaer Polytechnic Institute (RPI), recently published a MATLAB-based software package for fitting of isotherm data from batch experiments using a variety of SDM-type formalisms for ion exchange and multimodal chromatography using either single component or multicomponent formalisms. The software package is freely available to any interested parties, industrial or academic groups, and can be downloaded or cloned using the link to the IsoFit github repo.
The adsorption models available in the software are listed below, with more information provided in the document “Description of isotherm models” on the github repo.
- SMA (recommended for IEX or multimodal cation exchange)
- SMA with Mollerup extension (recommended for IEX or multimodal anion/cation exchange)
- Ottens (Nfor/Ottens 2010 JChromA, for multimodal anion/cation exchange)
- Ottens with with Mollerup extension (for multimodal anion/cation exchange)
- SMAHIC hybrid (HIC portion from Wang/Hubbuch 2016 JChromA, for multimodal anion/cation exchange)
- SMAHIC hybrid with Mollerup extension (for multimodal anion/cation exchange)
All of these isotherm models are available with empirical pH extensions on the linear parameters: charge as a linear function of pH and Keq as an exponential function of pH. All models can be used in multicomponent format and there are options included to reduce the isotherm parameter space which can quickly blow up when using multiple proteins, especially for the multimodal isotherms.
We are certainly always open to adding new isotherm models, particularly for other modes of operation such as affinity or HIC. Suggestions are always appreciated.
With regards to numerical methods: MATLAB fsolve (nonlinear function solver) is used to solve the equations, in implicit form, for the equilibrium solid phase concentrations. Parallel processing may be enabled for expediency (each thread handling one point on the isotherm). MATLAB ga (genetic algorithm) is used to optimize the isotherm parameter set by reducing the formulated NRMSE objective.
An excel sheet template is provided which is to be filled out with the experimental batch data, model selection options, and isotherm parameter ranges for the estimation routine.
We are working on several publications using this software which will be available later this year.
Feel free to contact me here or at alters9595@gmail.com for any questions, issues, features requests, or potential collaboration.
Thanks,
Scott Altern
Cramer Group