Implementing model-based real-time monitoring in biopharmaceutical production represents a key advancement toward Quality by Design and provides the foundation for model-predictive control. Data-driven models are capable of making predictions for monoclonal antibody (mAb), double-stranded DNA, host cell protein, and high molecular weight impurity concentrations during elution from a Protein A chromatography capture step. Ensuring explainability of machine learning models and applying efficient experimental design are key to enhancing modern process control.

We present a physics-grounded binding isotherm for ion-exchange chromatography derived from electrical double layer (EDL) theory. Incorporating nonlinear screening, ion valency, and crowding, the model achieves predictive elution from minimal calibration. Its parameters map directly to physical properties, enhancing interpretability over existing models. Calibrated on a single dataset, we validate predictions across gradients, step elutions, salt types, and load densities with model proteins, and highlight the effect of divalent ions.

The development of effective shape descriptors is essential for representing the unique geometric features of protein surfaces. In this work, we introduce a novel approach based on spectral geometry: the Laplace-Beltrami spectrum. By encoding protein structures as fixed-length feature vectors derived purely from geometric properties, shape descriptors aim to facilitate the a priori prediction of protein separation, reducing modeling efforts, and accelerating chromatographic process design.

This work focuses on improving flowthrough multimodal chromatography for broader applications in biomolecule purification. By refining operational parameters and understanding key interactions, we aim to enhance process efficiency, product recovery, and impurity removal. The findings support more robust and scalable workflows, making this technique a valuable tool in both research and industrial settings. This approach highlights practical considerations for optimising chromatography without compromising performance or throughput.

The demand for Lentiviral Vector (LV) drug substance is increasing. However, primary capture using convective anion-exchange chromatography remains a significant manufacturing challenge. This presentation explores the factors influencing elution heterogeneity of lentiviral vectors during anion exchange chromatography.

This presentation will delve into how alternative purification strategies are being explored to meet the demand for higher yields, improved vector quality, and shorter processing times. We will discuss the implementation of novel adsorbent materials and non-conventional operational modes, with a focus on lentiviral vectors and extracellular vesicles.

Purification of bionanoparticles (BNPs), such as viruses, virus-like particles, and extracellular vesicles, is challenging due to their complexity, large size, and the presence of similar impurities. Resin-based packed-bed chromatography, though widely used, often suffers from low productivity due to slow mass transfer and limited binding capacity, as BNPs typically bind only to the bead’s outer surface. Non-woven fiber-based convective media provides a promising alternative, offering a larger accessible surface area for efficient purification in both flowthrough and bind-elute modes. This approach overcomes key limitations, enhancing productivity and BNP purity while maintaining the possibility for easy scalability and enabling sterile processing.

Viral capsomere–nucleic acid interactions introduce unique challenges to downstream processing and in vitro assembly of virus-like particles, as well as in vitro loading of viral vectors. These interactions affect purification efficiency, product quality, and scalability. By identifying key drivers of nucleic acid association, we present strategies that improve recovery and enable in vitro vector loading, where capsids are assembled cell-free from structural proteins and genetic cargo.

Novel peptide-based affinity ligands enable serotype-agnostic purification of AAV vectors and monoclonal antibodies, delivering yields up to 99% with 230-400-fold host cell protein reduction. Machine learning-guided Bayesian optimisation accelerates chromatography parameter development while preserving transduction activity. Advanced process analytical technologies, including quantum dot-enabled biosensors and electrochemical platforms, provide real-time monitoring of critical quality attributes. These innovations transform viral vector and mAb biomanufacturing through improved efficiency, scalability, and product quality.

As demand for rAAV gene therapies grows, continuous downstream processing (DSP) is key to achieving scalable, cost-effective, and safe manufacturing. This work integrates mechanistic modelling, digital twins, and Process Analytical Technology (PAT) to enable real-time monitoring, simulation, and optimisation. The approach improves process efficiency and consistency, supporting faster production and robust quality control—essential for meeting clinical and commercial needs in a rapidly evolving gene therapy landscape.

Hexameric IgGs are a class of complex biologics that offers unique pharmacokinetics and pharmacodynamic properties due to their high valency (up to 12 antigen binding sites), molecular weight (~900 kDa), or hydrodynamic radius. Despite their attractive profile for certain therapeutic indications, technical challenges in purification and analytics hinder their broader application. This talk aims to demystify some of the challenges by providing comprehensive molecule analysis/illustration and a couple of successful examples where key problems are identified and resolved. It would be of some guidance or introduction for those who are interested in making large and complex biologics for drug discovery and development.

Secretory immunoglobulin A (sIgA) is emerging as a powerful tool for passive immunisation. Current affinity resins suffer from co-purification of product-related impurities and severe mass transfer limitations. A novel resin was developed by immobilising a Streptococcus pneumoniae surface protein as an affinity ligand, targeting the secretory component of sIgA, on a macroporous resin with appropriate pore size. Data will be presented on how ligand design and resin architecture influence chromatographic performance. Additional topics include: affinity chromatography, secretory antibodies, resin development.