Differential Evolution for the Optimization of DMSO-Free Cryoprotectants: Influence of Control Parameters

Abstract

This study presents the influence of control parameters including population (NP) size, mutation factor (F), crossover (Cr), and four types of differential evolution (DE) algorithms including random, best, local-to-best, and local-to-best with self-adaptive (SA) modification for the purpose of optimizing the compositions of dimethylsufloxide (DMSO)-free cryoprotectants. Post-thaw recovery of Jurkat cells cryopreserved with two DMSO-free cryoprotectants at a cooling rate of 1 °C/min displayed a nonlinear, four-dimensional structure with multiple saddle nodes, which was a suitable training model to tune the control parameters and select the most appropriate type of differential evolution algorithm. Self-adaptive modification presented better performance in terms of optimization accuracy and sensitivity of mutation factor and crossover among the four different types of algorithms tested. Specifically, the classical type of differential evolution algorithm exhibited a wide acceptance to mutation factor and crossover. The optimization performance is more sensitive to mutation than crossover and the optimization accuracy is proportional to the population size. Increasing population size also reduces the sensitivity of the algorithm to the value of the mutation factor and crossover. The analysis of optimization accuracy and convergence speed suggests larger population size with F > 0.7 and Cr > 0.3 are well suited for use with cryopreservation optimization purposes. The tuned differential evolution algorithm is validated through finding global maximums of other two DMSO-free cryoprotectant formulation datasets. The results of these studies can be used to help more efficiently determine the optimal composition of multicomponent DMSO-free cryoprotectants in the future.

Cryopreservation of Human iPS Cell Aggregates in a DMSO-Free Solution—An Optimization and Comparative Study

“…Emergent population staying constant for two generations was defined as the criterion of convergence for the DE algorithm. This criterion has been validated in our recent study (Pi et al, 2019a) to ensure ≥95% accuracy of predicting the global optimum.…”

“…Compared to our previous DE algorithm studies (Pollock et al, 2017;Pi et al, 2019a), this study optimized four variable components instead of the former three variables, which resulted in six times as many possible vectors in the parameter space than before. However, the population size of each generation (i.e., 10) was smaller than what had been commonly used before (i.e., 13-27).…”

“…A DE algorithm with the basic mutation strategy (DE/rand/1/bin) (Storn and Price, 1997;Pollock et al, 2017;Pi et al, 2019a) was used to rapidly optimize the composition of the DMSO-free freezing solution for hiPSCs based on the functional metric of post-thaw reattachment rate. Briefly, the DE algorithm utilizes stochastic direct search to randomly generate an initial group (Generation 0) of sample parameters (i.e., concentrations of DMSO-free CPA molecules) from the population spanning the entire parameter space.…”

Machine learning and hypothesis driven optimization of bull semen cryopreservation media

“…al. 19 for strategy 1. The algorithm was completed, and convergence was achieved when post-thaw motility between multiple generations began to plateau.…”

Machine learning and hypothesis driven optimization of bull semen cryopreservation media

“…Recently machine learning was used to optimize a cell culture medium for T cells 16 and to design an integral membrane channel rhodopsin for efficient eukaryotic expression and plasma membrane localization 17 . Similarly, a differential evolution algorithm was used to optimize cryopreservation conditions for Jurkat cells and mesenchymal stem cells 18 , 19 . In these experimental designs protocols or medium are optimized via information obtained through physical experiments guided by the algorithm.…”