Dynamic Light Scattering (DLS) has been a well-established technique to estimate the average size of the aggregate or to evaluate the stability of the sample. In this study, we have discussed one more prominent solution that can be achieved by DLS, which is calculating the composition of a mixture. Most researchers use size exclusion chromatography to determine the composition of components in a mixture which is time-consuming. We have demonstrated a technique with which we can determine the oligomer composition of a sample with the help of its DLS data. The technique is based upon a machine learning algorithm and the implication of a regression model to predict the oligomers in the monoclonal antibodies. We were able to predict the percentage of monomer, dimer etc with an absolute error of less than 14%. This proposed technique could be economical and an alternative way for rapid estimation of oligomers in the sample.

Prediction of percentage aggregate in a sample from DLS data using a machine learning algorithm

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.