Leveraging Data Science for Enhanced Protein Expression

Teaching AI models’ protein prediction can be slow and cumbersome, where design cycles (design, make, test, and analyse) often take months and require specialist expertise across multiple teams. The data that research teams have is often not suitable for training models because it is generated to support a specific project and is not standardised. Using IVTT (cell-free protein expression), integrated liquid handlers, and automated data analysis, we streamline the process and generate fit-for-purpose data. In 7 days, we go from DNA to analysed data.

AstraZeneca’s Protein Sciences & Analytics is building the lab of the future based on an automation-first operating model and AI-ready data capture. The enterprise-scale blueprint is grounded in harmonized ways of working, small- and large-scale HT expression platforms, and integrated data pipelines. The ongoing transformation has already shown measurable impact on DMTA timelines and first-pass success and aims to be transferable across modalities and sites—offering a clear path to enable faster, data-driven decisions.

High-throughput biophysical data generation is emerging as the missing link in AI-driven protein design. We present a platform for scalable production of diverse drug target proteins combined with multi-modal characterisation, including nano-DSF, BLI, SPR, and FIDA. Our approach captures protein stability, aggregation, and binding interactions—particularly for VHHs—at high granularity. By generating paired datasets that link sequence, expression, and biophysical behaviour, we expand both the diversity and resolution of training data. This data-centric strategy enables more robust, generalisable, and predictive AI models, helping to unlock design capabilities in data-sparse and challenging regions of protein space.

Membrane proteins are important drug targets, but their expression and purification can be challenging and delay drug development. Recent advances in computational protein design enabled the engineering of water soluble, stable analogues that accurately mimic epitopes of their native counterparts. This presentation will explain the development and validation of membrane protein analogues and highlight their application and benefits through an example of nanobody discovery and characterisation.

Improving protein production is critical in today’s rapidly advancing protein technology landscape. Using experimental data, we developed a codon-efficiency metric that correlates with the levels of native and recombinant proteins in Escherichia coli. Codon content influences protein expression more than mRNA folding, except in the first few codons, by modulating translation and mRNA degradation. An A-rich, G-poor base composition in the first six codons enhances expression and mRNA stability. We integrated these findings into a sequence optimisation algorithm that predicts the expression of a given DNA sequence in E. coli and proposes strategies for high-yield protein production.

Protein engineering increasingly demands methods that can efficiently navigate vast sequence spaces to identify variants with desired properties. We developed an integrated deep mutational learning pipeline combining high-throughput experimental fitness landscapes with protein language models to engineer proteins across two distinct applications: enhancing myoglobin peroxidase activity through electron-hole hopping pathways, and tuning the sensitivity and dynamic range of HucR-based biosensors. In both systems, neural network models trained on sort-seq and EP-Seq data accurately predicted improved variants, achieving up to 100-fold enrichment over random mutagenesis. Our results establish a generalisable framework for accelerating precision protein engineering.

We present our experience expressing over 80,000 AI-designed proteins using cell-free systems, systematically evaluating expression tags, codon optimisation strategies, and system variants. Applying our automated cloud laboratory for high-throughput protein validation we describe integrating these workflows into this fully automated platform capable of expressing arbitrary customer proteins at scales from sub-microgram to milligram quantities.

Leveraging computational protein design, we developed an integrated pipeline to generate miniprotein binders against targets of interest. A modular Golden Gate-compatible pJL1 plasmid and optimised cell-free protein synthesis (CFPS) enabled rapid and cost-effective in-house production. Combined with TwinStrep-mediated capture, SPR-based hit screening directly from crude extracts enabled functional validation without intermediate purification. This integrated approach establishes a transferable framework for accelerating binder discovery through rapid, economical production and purification-free screening.

The development of therapeutic antibodies requires reliable and reproducible production platforms. Chinese hamster ovary (CHO) cells are widely used for biopharmaceutical manufacturing; however, establishing stable producer cell lines remains a challenge. This project aims to establish a stable CHO-K1 cell line producing a humanized anti-IL7R antibody, developed for investigation in the context of pediatric acute lymphoblastic leukemia (ALL), through CRISPR/Cas9-mediated site-specific integration. The strategy targets the C12orf35 locus and employs homology-directed repair (HDR) to introduce the antibody expression cassette into a defined genomic region. The experimental approach encompasses guide RNA design, CRISPR/Cas9 construct development, evaluation of plasmid delivery, and molecular validation of targeted integration. This work seeks to establish a controlled platform for stable anti-IL7R antibody production, contributing to the development of reproducible cell-based systems for therapeutic antibody manufacturing.

Membrane protein-expressing cell lines are essential for antibody discovery, whole-cell phage display, hit screening, and bioassays. However, generating stable cell lines for diverse membrane proteins often requires tailored approaches. Here, we evaluated the DirectedLuck transposase for research cell line development. The system accelerated generation of high-expressing stable pools and supported sustained transgene expression, providing an efficient platform for streamlined and robust cell line development.

Heterologous protein expression remains a bottleneck in biomanufacturing, often hindered by trial-and-error workflows and unpredictable yields. This presentation introduces a computational platform integrating mechanistic biophysical models with deep learning to design optimised genetic constructs. By modelling translation dynamics, ribosome kinetics, and mRNA structure across diverse hosts, we move beyond legacy codon optimisation. We will demonstrate how this predictive engineering approach maximises protein titres in microbial, mammalian, plant, and insect systems while significantly reducing R&D cycles. Attendees will learn how integrating data science into gene design transforms protein production from an experimental burden into a precise, scalable digital workflow.

REM-BAC, our novel baculovirus-based platform for generating stable mammalian producer lines, represents a versatile, cell type independent, cost-effective platform incorporating several key advantages of the conventional BacMam technology: preferable biosafety profile, efficient transduction of various cell types, delivery of large DNA cargos. In combination with site-specific genome integration of large transgenes with customisable expression levels and integration of transgenes without leaving viral footprints, it ensures scalable manufacturing of next-generation biopharmaceuticals.