Currently, there is a high demand for fast to market vaccines, of which the global COVID-19 pandemic in 2020 and 2021 is a prime example, and accelerated process development is needed to meet this demand. Downstream processing is the main bottleneck during process development for protein-based vaccines, especially the purification step. This is due to the main method used for purification, namely chromatography. Due to the versatility and specificity of chromatography, vastly different parameters need to be tested to find optimal separation conditions. This leads resource-intensive experimental screenings and, consequently, long process development times. To optimize process development, we are developing a computational tool that can predict chromatographic behavior of proteins based on information retrieved from the atomic structure. More specifically, we apply Quantitative Structure Property Relationship (QSPR) models, where protein descriptors are calculated based on structural properties. Development of these models requires proprietary software or many different open source tools that focus on specific aspects, such as solely the calculation of protein electrostatics, surface/volume, or hydrophobicity. As a result, one needs to combine information from different sources. This can prove to be a difficult and time consuming task, and subsequently discourages inexperienced users. The work we will present includes an open-source Python project capable of calculating different descriptors in one environment. This tool provides easy to read outputs that can be used as input for machine learning models with minimal to no handling required. Our goal is to make QSPR available for a more general audience by lowering the required technical knowledge of such approaches. Via this route, QSPR models become more straightforward for widespread implementation and can contribute to the reduction of process development time.

Protein quantitative structure-property relationships (QSPR) for improved chromatographic separation

ACS Spring 2022

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