The Xu Lab builds Raman technologies for sorting and profiling individual cells, and pairs them with machine learning to turn spectra into diagnoses. Our work spans rapid culture-free identification of pathogens and antimicrobial resistance, label-free profiling of immune cells for cancer, autoimmune and chronic disease, and single-cell routes to sustainable biotechnology.

Jiabao Xu is a Lecturer (Assistant Professor) in Biomedical Engineering at the University of Glasgow, where she leads a group working on single-cell Raman spectroscopy and machine learning. She read Chemistry at Imperial College London, took a DPhil (PhD) in Engineering Science at Oxford, and spent 2019 as a visiting scholar in Chemistry and Engineering at Boston University before returning to Oxford as a postdoctoral research scientist. She moved to Glasgow in 2023.
Her work asks what a single living cell is doing chemically, without staining it or killing it, and what becomes possible once that can be read at scale: identifying pathogens and their resistance without culture, telling immune cells apart by their metabolism, and sorting engineered microbes by what they actually produce. She holds a Fellowship of Recognising Excellence in Teaching from the University of Glasgow, and is a Visiting Professor at Shanghai East Hospital, Tongji University.
The instruments, chemistries and computational methods the lab develops.
Raman-Activated Cell Sorting (RACS), Raman Stable Isotope Probing (Raman-SIP), label-free Raman imaging.
Machine learning and deep learning pipelines for disease classification, multimodal learning across Raman spectra, omics, images and clinical metadata, denoising autoencoders, explainable AI.
The biological and clinical questions we put those methods to.
Single-cell pathogen and immune profiling, rapid pathogen identification, antimicrobial resistance profiling, culture-free diagnostics.
Raman-based PBMC profiling, label-free diagnostics for cancer, autoimmune disorders such as SLE and psoriatic arthritis, and chronic disease.
Single-cell phenotyping of engineered microbes, metabolic engineering and promoter selection for bioproduction, engineered photosynthesis and CO2 fixation, whole-cell biosensors, and microbial routes to bioplastics and waste valorisation.
PEEN turns an expanded hydrogel into an active plasmonic nanoreactor, so that a standard benchtop Raman spectrometer can resolve metabolic activity at roughly 25 nm. The post is a chemistry post: the resolution comes from the sample chemistry rather than from the optics.
DECODE-FATIGUE follows an unexpected finding: in psoriatic arthritis the extreme fatigue phenotype is biologically non-inflammatory, and carries an oxidative stress signature strongest in CD8+ T cells. The post takes that signature apart and looks for something druggable in it, so it is an immunology and cell biology post rather than a spectroscopy one.