An innovative image-based classification tool was developed to address subvisible particle challenges in biopharmaceuticals. By combining feature extraction with machine learning, it delivers over 97% accuracy in distinguishing silicone oil from other particles. The approach provides a cost-effective alternative to complex analytics, strengthening quality control and process reliability. Faster, consistent monitoring supports regulatory compliance, safeguards product integrity, and enhances operational efficiency across biopharmaceutical manufacturing and development.

This presentation highlights emerging methodologies and technologies that expand the resolution, sensitivity, and throughput of O-linked glycomics and glycoproteomics, including the use of mucinases that now are available to improve the glycopeptide generation of the highly O-linked mucin domain frequently found on O-glycoproteins. By redefining analytical capabilities, these next-generation approaches enable deeper biological understanding and support the design, optimisation, and quality control of innovative biologics and complex therapeutic modalities.

Glycosylation is a common PTM impacting protein function, disease, and therapeutic efficacy. Its characterisation is essential, as glycosylation critically influences stability and pharmacokinetics of biopharmaceuticals. Multi-level mass spectrometry approaches—ranging from released glycans, glycopeptides and intact proteins—provide complementary information on different layers of heterogeneity (e.g., micro and macroheterogeneity). Integration of these data enables comprehensive glycosylation profiling, facilitating a deeper understanding of co-occurring modifications and ensuring reliable assessment of glycosylation.

Glycan profiles of therapeutic proteins are key molecular attributes that affect biological function and efficacy. The control of the profile during bioprocessing is difficult because of the multiplicity of factors that affect intracellular glycosylation. An alternative approach is to modify the glycan profile during downstream processing, after purification from the bioreactor. Alternative methods will be explored that include selecting specific glycoforms or enzymatically modifying the glycans on the protein.

Lipases are Host Cell Proteins (HCPs) that co-purify with biological drug substances (DS) and degrade excipients such as Polysorbates. LC-MS can identify and quantify Lipases which can guide process development to revise their purification to remove problematic HCPs. Absolute and Relative LC-MS quantification approaches will be compared with case studies focusing on Lipase quantification in DS samples.

The complexity of lipid nanoparticles (LNPs) for delivery of nucleic acids presents a developability challenge. Fit-for-purpose and complementary analytical tools are required to design successful formulations, and to inform robust production of stable, safe, and efficacious products. This talk will cover some case studies involving novel and repurposed biophysical techniques for mRNA and LNP characterisation with the emphasis on approaches with stability-indicating potential.

Pulmonary drug delivery (PDD) for biologics is an emerging focus in drug delivery research. By exploiting the lungs’ large surface area, rich vascularization, and rapid absorption, it offers a promising route for peptides, proteins, and nucleic acids. Advances have deepened understanding of this therapy, yet key challenges persist: maintaining molecular stability during formulation and aerosolization, minimising immunogenicity, and creating effective systems to protect these sensitive molecules while ensuring bioavailability.

Post-translational modifications (PTMs) are critical for regulating protein function but challenging to analyse due to their complexity. This project aims to create an automated pipeline for high-throughput PTM screening, leveraging LC-MSE (a data-independent acquisition method), alongside customised data processing tools. The workflow automates the integration of spectral analysis, validation, and filtering of PTM fragments based on peptide attributes—ensuring accurate, reproducible results while reducing false-positive errors and minimising manual interpretation.

N-glycosylation strongly affects monoclonal antibody (mAb) quality and efficacy. We applied machine learning (ML) to predict N-glycan abundances in CHO cell fed-batch cultures under 12 media conditions. From mass spectrometry data, ML models reduced 167 peaks to 18 predictive features. Integrating simulated annealing enabled media design that improved titer while reducing mannosylation, illustrating how ML-guided monitoring and simulation can accelerate process optimisation in mAb biomanufacturing.

Hybrid modelling and machine learning techniques are rapidly advancing in process development to optimise performance and accelerate timelines. These techniques primarily focus on offline, iterative modelling, unlike manufacturing applications that require real-time decision support. The presentation will showcase the application of hybrid modelling and transfer learning from the manufacturing perspective and how advanced analytical techniques like Raman and mass spectrometry can enhance knowledge and improve online process monitoring and control.