Programme

DNA methylation is an epigenetic mark and large-scale alterations in DNA methylation patterns are observed in development, ageing and diseases such as cancer. Previous mathematical models have aided in understanding methylation patterns, but the computational expense associated with these studies have limited their use to the study of small-scale patterns. During this talk, I will discuss using cluster mean-field models to quickly, and accurately, predict statistical properties associated with large-scale DNA methylation patterns. Much of my talk is based on the publication https://royalsocietypublishing.org/doi/10.1098/rsif.2021.0707

My past academic research focused on using mathematical and computational models to investigate the growth and collective migration of cancer cell populations. The main focus of that research was to look at heterogeneous populations – where subtypes of cancer cells had different abilities – and how that can drive metastasis and spread of the cancer. This work was mathematically rigorous and explored methods for modelling these phenomena, however, was not a particularly data-driven approach.

My current position, as a quantitative systems pharmacology (QSP) scientist at Certara, has allowed me to continue using similar mathematical methods in real world applications. At Certara, we use QSP modeling to predict how new treatments may work in humans. Within these models, we use preclinical and clinical datasets to calibrate and validate the modelling techniques. These models can then be used to support the design and approval of further clinical trials.

My talk will briefly cover some of the differences and similarities between these theoretical and data-driven research approaches. Both approaches are essential—working together, they advance cancer research and more generally, biomedical research.

In von Hippel–Lindau (VHL) disease, patients develop multiple independent kidney tumours that are monitored by serial imaging over years before surgery, generating dense longitudinal growth data that is typically unavailable in sporadic cancers. This makes VHL a rare natural laboratory for asking what determines how fast a tumour grows. Analysing volumetric growth rates across the cohort, we find that growth is not constant but accelerates approximately exponentially with tumour size: a size-dependence we term growth “steepness", which cannot be explained by geometric scaling alone. Because most determinants act as modifiers of this size-dependence rather than as size-independent effects, they become visible only once growth is modelled with explicit size-by-covariate interactions. Doing so brings hidden determinants into focus: spanning patient germline background, tumour-intrinsic drivers, and the microenvironment - several ascertainable before resection - and reveals that steepness itself converges within patients, identifying it as a host-anchored property rather than a tumour-cell-autonomous one. Strikingly, the average onset of acceleration coincides with the ~3 cm mark at which surgery is recommended, hinting that this clinical rule of thumb has mirrored a mathematically-tractable transition all along. In this talk, I will show how a single, mathematically-informed change in framing turned an apparent null result into a map of what drives tumour growth - reframing the clinical question from which tumours will turn aggressive to when.

My career has been shaped by a fascination with understanding complex biological systems through mathematical and computational approaches. A central focus of my research is understanding how cancer cells interact with their surroundings. The tumour microenvironment is a highly dynamic and spatially organised ecosystem in which interactions between cancer cells and the extracellular matrix (ECM) can drive tumour invasion and metastatic progression. Capturing these processes requires approaches that bridge biological complexity with quantitative and computational thinking.

In this talk, I will present my work on developing 3D individual-based models of the tumour microenvironment that explicitly represent cancer cells, ECM structures, and other tissue components. These models allow us to investigate how cell movement, ECM remodelling, and local interactions collectively shape tumour behaviour. I will also discuss the translation of this work into PhysiMeSS, an extension of the widely used PhysiCell platform developed through collaboration with researchers across the international community. This has enabled a framework originally developed within my own research programme to become a shared community resource, demonstrating the power of collaborative and community-driven computational science.

Alongside the science, I will reflect on my journey into computational biology and the many unexpected twists and turns that have shaped my career. Navigating an interdisciplinary research path while balancing the complexities of life beyond academia has influenced not only the questions I ask, but also how I approach collaboration, leadership, and mentoring. I will share some of the lessons learned along the way, highlighting the importance of supportive networks, strong mentors, and diverse perspectives in tackling some of the most challenging problems in modern biomedical research.

In recent years many institutions have adopted research culture strategies and action plans, influenced at least in part by the inclusion of new criteria in the Research Evaluation Framework. Equality, Diversity and Inclusion principles are often core to these plans. But although the “research culture” terminology is relatively recent, the EDI challenges in research environments and efforts to address them are not new. In this talk I will give some personal perspectives on this landscape and in particular, how the picture can become even more complicated when multiple disciplines, with different default cultures, collide.

Background

Tumour-infiltrating lymphocytes (TILs) are clinically informative in breast cancer, but conventional assessment is limited by observer variability and spatial heterogeneity. ECTIL (label-efficient computational tumour-infiltrating lymphocyte assessment) is a recently validated computational pathology model that estimates stromal TIL abundance from routine H&E whole-slide images (1). Using three complementary breast cancer cohorts, CPTAC-BRCA, TCGA-BRCA and TransNEO, we examined the biological, prognostic and treatment-response relevance of ECTIL. We investigated what biological information ECTIL captures, whether its correlation with immune-related signalling is consistent across cohorts, and whether ECTIL provides prognostic information beyond established transcriptomic immune signatures.

Methods

CPTAC-BRCA (2) was used as the primary molecular characterization cohort. In 110 patients with matched RNA and 273 eligible H&E slides, the ECTIL-compatible workflow included foreground segmentation, normalization to 0.5 µm/pixel, extraction of non-overlapping 512×512-pixel tiles, RetCCL feature extraction and frozen ECTIL-TCGA inference. For patients with multiple eligible slides, patient-level ECTIL was calculated as the tile-count-weighted mean.

Before transcriptome-wide analysis, ECTIL was tested against six prespecified immune RNA measures representing distinct biological axes: cytolytic activity (CYT), IFNG-6, CD8/T-cell core, T-cell-inflamed positive-17, B-cell/humoral-5 and ESTIMATE ImmuneScore. Spearman correlations were estimated with 10,000-bootstrap confidence intervals and Benjamini-Hochberg (BH) correction across the six measures. Transcriptome-wide analysis was then used to characterize gene-level relationships with ECTIL, with covariate-adjusted HC3 regression as a complementary analysis; gene-level significance was defined at BH FDR<0.05. Nested Elastic Net stability selection was subsequently used for compact RNA-score development; no clinical outcome was used during this process.

TCGA-BRCA (3, 4) provided independent RNA prognostic validation and a separate training aware image analysis. Frozen RNA scores were calculated in 1,058 primary-tumour cases. ECTIL was successfully generated for 1,122 processable H&E slides from 1,051 patients; seven slides lacking essential microns-per-pixel metadata were excluded for technical reasons. After the primary TCGA package was frozen, released ECTIL train, validation and test files were used to reconstruct the original labelled TCGA cohort of 356 slides from 342 patients and identify 710 image-RNA patients outside that cohort. Post-freeze sensitivity analyses repeated ECTIL-RNA correlations in these patients and examined receptor-phenotype consistency. Progression-free interval (PFI) was the primary clinical endpoint; DSS and OS were secondary and DFI exploratory. Cox models used a frozen clinical adjustment backbone comprising age, ordinal pathologic stage, and receptor-defined clinical phenotype. To assess complementary prognostic information, PFI models compared the frozen clinical backbone alone, clinical+ECTIL, clinical+RNA and joint clinical+ECTIL+RNA models. Incremental prognostic value of ECTIL beyond each of the six established RNA signatures, and RNA beyond ECTIL, was assessed using nested likelihood ratio tests with BH correction across the six-signature family (BH-adjusted p values reported as q values). ECTIL×RNA interactions were evaluated separately.

TransNEO (5) provided independent neoadjuvant-response validation. ECTIL was generated for 203 H&E slides from 203 patients; 147 patients had matched ECTIL, RNA and response data. Associations with pCR and continuous residual cancer burden (RCB) were tested and joint ECTIL-RNA models assessed complementary response information.

Results

Across cohorts, patient-level ECTIL ranged from 0.070–0.557 in CPTAC-BRCA, 0.034–0.617 in TCGA-BRCA and 0.051–0.604 TransNEO image cohort. In CPTAC-BRCA, ECTIL was positively associated with all six prespecified immune RNA measures. Five remained significant after multiplicity correction: IFNG-6 (ρ=0.343, q=0.002), T-cell-inflamed positive-17 (ρ=0.277, q=0.010), CYT (ρ=0.261, q=0.012), CD8/T-cell core (ρ=0.251, q=0.012) and ESTIMATE ImmuneScore (ρ=0.223, q=0.023); B-cell/humoral-5 was positive but weaker (ρ=0.149, q=0.121). Transcriptome-wide analysis identified 7,061 genes at FDR<0.05, of which 1,622 were significant in both the primary and age-, stage- and TNBC-adjusted analyses. Stability development yielded a compact nine-gene ECTIL RNA score (BATF2, IFNG, LAG3, TNFRSF13C, KIR2DL4, PLA2G2D, HMGB2, HSPD1 and CD320), with repeated held-out internal CPTAC correlation ρ=0.618; in independent TCGA RNA validation, the score was associated with longer adjusted PFI (HR per SD=0.67, 95% CI 0.53–0.84; p<0.001).

In the 1,051 patient TCGA image-RNA cohort, ECTIL was positively correlated with all six established immune RNA measures and the nine-gene score (all BH q<0.05). In the 710 patients outside the original labelled ECTIL TCGA cohort, ECTIL remained correlated with all six established RNA measures (ρ=0.235–0.317; all BH q<0.05) and the nine-gene score (ρ=0.427, q=4.83×10-32). The ECTIL–nine-gene score relationship remained positive across major receptor phenotypes, including TNBC, with no evidence of heterogeneity in the full cohort. In fully adjusted joint PFI models using a common 895-patient, 96-event analysis set, higher ECTIL remained associated with lower progression risk when modelled alongside each established RNA signature, with ECTIL HRs per SD ranging from 0.61 to 0.65. ECTIL added significant prognostic information beyond all six established RNA signatures after multiplicity correction (BH q=0.0017–0.0029). In contrast, none of the six established RNA signatures added significant prognostic information beyond ECTIL after correction, and no ECTIL×RNA interaction was significant across the six-signature family. This incremental-value pattern persisted in the post-freeze TCGA sensitivity restricted to patients outside the original labelled ECTIL cohort (n=599, 60 PFI events): ECTIL added significant prognostic information beyond each established RNA signature after BH correction, whereas reciprocal RNA-beyond-ECTIL and interaction tests were not significant after correction.

In the TransNEO response validation (n=147; 38 pCR), higher ECTIL was associated with increased odds of pCR (adjusted OR per SD=1.73, 95% CI 1.18–2.54; p=0.005) and lower continuous RCB (adjusted β per SD=−0.33, 95% CI −0.52 to −0.14; p<0.001). None of the six established RNA signatures added significant pCR or RCB information beyond ECTIL after multiplicity correction; ECTIL retained incremental information beyond B-cell/humoral-5 (pCR q=0.036; RCB q=0.034).

Conclusions

ECTIL showed reproducible correlation with multiple established immune transcriptomic signatures, retained prognostic information beyond individual RNA immune measures, and independently associated with pathological treatment response. These findings support ECTIL as a biologically interpretable computational pathology biomarker complementary to bulk immune transcriptomics. The nine-gene score may provide a compact secondary RNA representation of the ECTIL-associated transcriptomic signal.

References

1. Schirris Y, Voorthuis R, Opdam M, Liefaard M, Sonke GS, Dackus G, et al. Label-efficient computational tumour infiltrating lymphocyte assessment in breast cancer (ECTIL): multicentre validation in 2340 patients with breast cancer. The Lancet Digital Health. 2025;7(11).
2. Krug K, Jaehnig EJ, Satpathy S, Blumenberg L, Karpova A, Anurag M, et al. Proteogenomic Landscape of Breast Cancer Tumorigenesis and Targeted Therapy. Cell. 2020;183(5):1436–56.e31.
3. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490(7418):61–70.
4. Liu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD, et al. An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics. Cell. 2018;173(2):400–16.e11.
5. Sammut SJ, Crispin-Ortuzar M, Chin SF, Provenzano E, Bardwell HA, Ma W, et al. Multi-omic machine learning predictor of breast cancer therapy response. Nature. 2022;601(7894):623–9.

Cancer cachexia is a systemic metabolic rewiring resulting due to advanced cancer in severe muscle and fat loss. It markedly reduces tolerance to treatment and contributes to two million deaths annually. Despite its clinical impact, cancer cachexia remains poorly characterised [1]. Positron Emission Tomography (PET) provides a whole-body functional readout of metabolism and is routinely acquired in cancer care [2]. We present an automated PET quantification pipeline to explore the underlying metabolic mechanisms of cachexia through characterisation of organ-level changes, inter-organ relationships, and cancer-type specific cachexia signatures.

Two large multicentre [¹⁸F]fluorodeoxyglucose PET datasets were collated: Kaiser Permanente (lung n=2,289, pancreatic n=572, colorectal n=305) and TRACERx (n=643 NSCLC). Kaiser Permanente patients were stratified by weight-loss trajectory relative to scan date. TRACERx patients were categorised based on body composition changes. Whole-body PET/CT scans were segmented using TotalSegmentator [3], with organ-level SUV adaptive thresholds and growth filters applied for primary and metastatic tumour segmentation to exclude physiologically implausible uptake values. Organ-level metabolic activity was normalised to aortic uptake, yielding Standardised Uptake Value Ratios (SUVR), which mitigate scanner and protocol variability across sites. Organ-level SUVR features were extracted across 70 anatomical structures. Univariate discrimination was assessed using Cliff’s delta effect sizes and the Mann-Whitney U test. Multi-organ classification models (XGBoost, Random Forest) were built using minimum Redundancy Maximum Relevance (mRMR) feature selection. Balanced accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC) were used to measure separation, and feature importances were used to derive metabolic fingerprints. Linear Discriminant Analysis (LDA) characterised metabolic differences across cancer types and demographic subgroups, alongside stratified repetitions of the primary analysis. Mutual information (MI) quantified altered inter-organ metabolic coupling networks, revealing systemic coordination beyond individual organ effects.

Individual organ variations were identified across the pan-cancer cohort in the Kaiser Permanente dataset. Most prominent was reduced uptake across several heart chambers in all three cancer types, also evident as a mild effect prior to weight loss. In colorectal and lung cancer, increased uptake in spleen and bone marrow indicated an inflammatory response. Across most organs, pancreatic cancer showed reduced PET uptake compared to the non-cachectic cohort.

LDA investigations into demographic variation demonstrated that while the severity of FDG uptake changes is influenced by age and sex, the overall trends are broadly consistent. Smoking status and comorbidities, however, had a substantial effect on PET presentation. MI demonstrated changes in muscle and cardiac inter-organ connectivity. The dynamic nature of cachexia was also evident, with several organs such as torso fat and kidney, showing significant uptake changes only at higher levels of weight loss.

Overall separation between cachectic and non-cachectic patients was achieved in Random Forest with 69.7 ± 2.2% balanced accuracy and AUROC of 0.772 ± 0.018 across the pan-cancer cohort.

TRACERx NSCLC analysis is at an earlier stage but has demonstrated trends consistent with the lung cancer Kaiser Permanente cohort, including reduced myocardium uptake at baseline and a similar increase in thyroid gland and skeletal muscle uptake.

This work demonstrates that PET, approached as a high-dimensional whole-body dataset rather than a lesion-level clinical readout, can reveal whole-body metabolic rewiring in cancer cachexia. Integrating this imaging layer with molecular and clinical data is a natural next step.

References

1. Ferrer et al. Cell. 2023
2. Ferrara et al. J Cachexia Sarcopenia Muscle 2024
3. Wasserthal et al. Radiology : AI 2023

Microscopy datasets are often spatially sparse, wherein relevant structures occupy only a small fraction of the total field of view (FOV), leaving large regions of background devoid of signal. This inherent inefficiency creates file sizes that are larger than needed, which increases the time needed for computational image data analysis and processing, and means unnecessarily large data volumes. In this work, a classical open-source method for content-aware spatial compaction of microscopy images (CASC) is reported that explicitly removes spatial redundancy by reorganising foreground objects into a new, smaller image. CASC combines adaptive intensity normalisation, statistical thresholding, morphological refinement, and connected-component analysis to isolate foreground structures. These structures are then extracted with contextual padding and repacked into a compact domain using a heuristic shelf-based spatial packing strategy. CASC intentionally destroys the spatial topology of the image but preserves pixel intensities exactly, retaining object-level information. The method achieves reductions in image area and background content while maintaining high object-level preservation of biological structures, with a reduction in file size of more than 390-fold shown in real biological image datasets.

Mutations in PIK3CA are frequently reported in solid cancers and cause a rare group of developmental overgrowth disorders termed PIK3CA-related overgrowth spectrum (PROS). Within the PROS patient population, low-flow vessels (veins, lymphatics, capillaries) are more commonly affected than fast-flow vessels (arteries), and it remains unclear why the penetrance of PIK3CA mutations varies across lineages. We hypothesise that signalling potential increases over the course of differentiation, driven by differential expression of regulatory subunits and other pathway components, and that this differential decoding of input signals underlies lineage-specific sensitivity to activating PIK3CA mutations. To address this, we are using iPSC-derived endothelial cell models and mass cytometry (CyTOF) to map signalling landscapes across pluripotent and vascular lineages, with and without activating PIK3CA mutations. As a foundation for this analysis, we are optimising our experimental workflow, including validating antibodies in our newly designed panel spanning lineage, signalling, and additional markers, to ensure any future findings reflect true biology rather than technical artefact. To date, of our 24-antibody panel (7 extracellular, 17 intracellular), 12 have been validated, 3 failed validation, and the remaining 9 require further testing. This work lays the groundwork for potential tissue-specific targeting of the PI3K pathway, with the aim of reducing systemic toxicity and improving treatment options for individuals with PIK3CA-driven disease.

The combined in vitro and in silico findings indicate that Irvingia wombolu leaf extract contains bioactive compounds capable of inhibiting α-glucosidase and may therefore contribute to the control of postprandial hyperglycaemia associated with type 2 diabetes mellitus. Among the identified compounds, linolenic acid showed the most promising inhibitory potential, followed by γ-tocopherol and α-tocopherol-β-D-mannoside. These findings support the potential of Irvingia wombolu as a source of novel antidiabetic agents and provide a basis for further studies involving compound isolation, in vivo evaluation, and drug development.

Breast cancer is a highly heterogeneous disease with several molecular subtypes: Luminal A, Luminal B, HER2+ve, and TNBC. These subtypes greatly vary in their molecular mechanisms, such as their sensitivity to hormonal signalling and growth factor pathways, thereby influencing prognoses, treatment strategies, and therapeutic responses. The intracrine steroid metabolic environment comprises the conversion of inactive precursors, such as DHEAS, to active forms of estrogens and androgens. The rearrangements of fluxes through estrogen and androgen production pathways play an important role in the manifestations of these subtypes and, hence, render distinctive metabolic behaviours.

Through this project, we aim to discern the dynamic variations in intracrine hormone production across breast cancer subtypes via kinetic modelling. Using transcriptome profiles from the TCGA-BRCA dataset, we evaluate relative changes in gene expression across subtypes and subsequently utilise these to parameterise enzyme activities across the estrogen-androgen metabolic network to simulate differential flux distributions. Through such a quantitative modelling framework for subtype-dependent steroidogenic rewiring, we aim to derive mechanistic links between transcriptional heterogeneity and metabolic phenotypes, elucidating each subtype’s unique intracrine environments.

Protein-protein interactions are integral to many key cellular processes, yet accurately predicting heterodimer structures remains challenging. Many state-of-the-art prediction workflows rely on multiple sequence alignments (MSAs), which are costly to generate, while others reduce MSA use but still retain complex folding modules. SimpleFold is an efficient MSA-free flow-matching model that combines protein language model (PLM) embeddings with a general-purpose transformer architecture, but has been evaluated only for monomer folding. In this work, we adapt SimpleFold for two-chain heterodimer prediction to test whether fine-tuning can extend its capabilities to complex prediction without altering its core architecture.

To train the model, we retrieved experimentally determined heterodimers from the Protein Data Bank (PDB), clustered complexes at 30% sequence identity, and generated train/validation splits. The resulting datasets contain 1,841 strict heterodimer representatives and 4,905 representatives in an expanded merged set. We modified the SimpleFold training pipeline to retain complete dimers within the token budget and support heterodimer-specific training and validation. We then created a held-out test set from post-cutoff PDB structures, restricted to biological assemblies with exactly two interacting protein chains and filtered against training homologs. We will use this framework to evaluate global structure and interface prediction relative to the base SimpleFold model.

Cell metabolism is regulated by mitochondria, organelles found in thousands of copies per cell. Mitochondria contain their own genome (mtDNA) which accumulates mutations throughout an individuals lifetime. However, the dynamics of mtDNA mutation accumulation are not well understood and are further complicated by differences in the effects of mutations found at different dosages (heteroplasmies) in single cells. We use mathematical simulations to investigate the processes that shape the distribution of mtDNA mutation frequencies observed in healthy tissue and cancer samples.

Pancreatic ductal adenocarcinoma (PDAC) is projected to become the second leading cause of cancer-related mortality, with minimal improvements in 5-year survival over the past decade. Although PDAC originates from premalignant lesions, only a minority advance to invasive malignancy. Distinguishing which lesions are likely to progress would transform enabling timely preventative surgical intervention, but this remains a major clinical challenge. Intraductal papillary mucinous neoplasms (IPMN) are a form of pre-cursor lesion associated with about 15% of PDAC cases and are of particular interest since they can be detected on imaging scans, creating an opportunity for early intervention. Determining which IPMNs are most likely to progress to PDAC would therefore be invaluable for preventing this aggressive and invasive cancer. In addition to the more well-characterized genetic mutations associated with PDAC onset, it has become evident that the tumour microenvironment (TME) plays an important role in shaping malignant progression.

To investigate the spatial relationship between the TME and progression from IPMN to PDAC, we are using 10x Genomics Xenium spatial transcriptomics platform to profile low-grade dysplasia IPMN, high-grade dysplasia IPMN, and IPMN-derived invasive PDAC across multiple regions within the same patients undergoing surgical resection at the Glasgow Royal Infirmary in the UK or at the Pancreas Institute of the University of Verona in Italy. Through analysis of spatial heterogeneity across stages of progression, we aim to delineate how local microenvironmental differences may influence epithelial lineage fate decisions into PDAC. Consistent with previous studies, we observe that stromal expansion and a reduction in CD8 T cells across progression. Importantly, however, low-grade IPMN lesions themselves show substantial heterogeneity in both epithelial and tumour microenvironmental cell gene programmes, even within the same patient. These findings motivate a deeper investigation of the structural and transcriptional changes across these low-grade lesions. Our findings will help understand the factors driving progression from early lesions to invasive cancer, supporting efforts in early PDAC detection and informing clinical strategies for surgical resection.

Background

Circulating proteins are attractive causal candidates and druggable targets in breast cancer, but observational protein–cancer associations are vulnerable to confounding and reverse causation. Whether specific plasma proteins causally influence breast cancer risk and whether such effects differ across intrinsic molecular subtypes remains undefined.

Methods

We performed two-sample Mendelian randomisation (MR) using cis-acting protein quantitative trait loci for 1,691 plasma proteins (deCODE, SomaScan v4; 5,890 genetic instruments) as exposures, and Breast Cancer Association Consortium (BCAC) summary statistics as outcomes, for overall breast cancer and five ER/HER2-defined intrinsic subtypes (luminal A, luminal B, luminal B HER2-negative, HER2-enriched, triple-negative). Effects were estimated with the TwoSampleMR framework (Wald ratio/inverse-variance weighted as primary; MR-Egger, weighted median and mode as sensitivity analyses), with Benjamini–Hochberg false discovery rate (FDR) control per outcome.

Results

Genetically predicted levels of 43 proteins were associated with overall breast cancer risk (FDR<0.05), 20 with luminal A and 4 with triple-negative disease. Eighteen proteins showed directionally consistent effects across multiple subtypes, whereas six reached significance in a single subtype, most strikingly RALB in triple-negative disease, indicating both breast-cancer-wide and subtype-restricted signals. Five proteins (TLR1, LAYN, ULK3, TNXB, CPNE1) were also found significant by other research groups who ran their own, separate MR studies on breast cancer.

Conclusion

Proteome-wide MR nominates candidate causal plasma proteins for breast cancer, several independently replicated, and distinguishes shared from subtype-specific effects with potential therapeutic-target relevance. Colocalisation and independent-cohort replication are ongoing to prioritise high-confidence targets.

Type 1 diabetes affects an estimated 9.5 million people worldwide. It is an autoimmune disease which leaves an individual unable to produce insulin and therefore their body has impaired blood glucose regulation. Standard treatment requires exogenous insulin supply and constant blood sugar monitoring. Islet transplants offer a means to reinstate insulin production. Isolated islets, from donor pancreases, are injected into the liver of the recipient via the hepatic portal vein where they engraft and secrete insulin. Currently, around 30 transplants are performed by the NHS per year but they face key limitations; multiple transplants are often required and lifelong immunosuppression is necessary, highlighting the need to better understand and optimise the process.

Here we developed a hierarchical Bayesian model to estimate dose-dependent islet survival from 18 mouse islet transplants across three dosing levels, allowing information to be shared across animals while maintaining separate dose-group estimates. The model identifies a notable increase in engraftment at the highest dose, despite substantial within-group variability. Cross-validation supported hierarchical information sharing over independent animal estimates, though dose group alone did not fully explain the variation, possibly reflecting the limited sample size or unmeasured covariates. We are now working to infer fragmentation and clumping rates of islets, with the longer-term aim of building a spatially-explicit model of engraftment dynamics.

Introduction

Colorectal cancer liver metastasis (CRCLM) is a leading cause of cancer-associated mortality. In the liver, disseminated cancer cells exhibit distinct histopathological growth patterns (HGP), of which the two most prevalent are desmoplastic (dHGP) and replacement HGP (rHGP). dHGP is characterised by the presence of a thick fibrotic ring separating metastatic cells from surrounding liver parenchyma, whereas rHGP is characterised by direct interaction between cancer cells and surrounding hepatocytes, allowing the co-option of pre-existing vasculature and gradual replacement of adjacent hepatocytes. rHGP is associated with resistance to anti-angiogenic therapy and worse prognosis. However, the mechanism by which different HGPs occur in the liver remains poorly understood. Therefore, we aim to characterise the immune microenvironment of dHGP and rHGP to better understand the role of immune cells in shaping the growth pattern of disseminated cancer cells within the liver.

Hypothesis

Immune regulatory cells are enriched in rHGP lesions, thus allowing disseminated cancer cells to co-opt pre-existing blood vessels and replace surrounding hepatocytes without triggering an inflammatory response.

Methods & Results

We utilised a publicly available single-cell RNA sequencing (scRNAseq) dataset to compare the immune population in rHGP (n = 3) and dHGP (n = 3) CRCLM patients. Principal analysis of the data revealed that dHGP lesions are enriched in lymphocytes, namely T cells (most of which are CD8+), NK cells, innate lymphoid cells, and plasma cells. rHGP lesions, on the other hand, are enriched in monocytes, macrophages, and B cells. We are planning to further characterise specific subtypes of immune cells that are differentially enriched in rHGP vs dHGP lesions. Additionally, as hepatic steatosis has recently been implicated in the development of rHGP lesions, we also plan to link the findings from our scRNA-seq analysis with the findings from our own Visium-HD dataset, which is comprised of six CRCLM patients, three of whom were clinically diagnosed with metabolic dysfunction-associated steatotic liver disease (MASLD) and three of whom were not.

Suicide outcomes, including suicidal ideation (SI) and suicidal behaviour (SB), are complex, highly heritable phenotypes with polygenic genetic architectures. Vitamin D (vitD) may influence mental health through roles in neurodevelopment, immune regulation, and neurotransmission, but its shared genetic basis with suicide phenotypes remains unclear. We investigated shared genetic architecture between vitD levels and SI/SB using an integrative genomic framework. GWAS summary statistics from large European-ancestry cohorts were analysed using conditional false discovery rate (cFDR), colocalisation, protein-protein interaction, and functional annotation approaches. Conjuction cFDR identified 53 pleiotropic SNPs for SI-vitD, including 29 lead SNPs (17 novel), and 53 for SB-vitD, including 38 lead SNPs (22 novel). Significant protein-protein interaction enrichment was observed for SI-vitD (P = 3.99 × 10⁻⁶) and SB-vitD (P = 0.0216). Colocalisation identified a shared causal locus on chromosome 3 for SI-vitD. Functional annotation highlighted brain cis-eQTL enrichment and CNS expression of prioritised genes, supporting evidence of shared genetic architecture between vitamin D and suicide outcome phenotypes and identifying biologically plausible genes and pathways relevant to neuropsychiatric vulnerability.

SPLICE is a data platform being developed at CRUK Scotland Institute to enhance data management and analysis in life sciences research. By combining metadata harmonisation, automated workflows and high-performance computing within a single platform, SPLICE supports end-to-end research processes across multi-omics, imaging and other experimental modalities. Built on FAIR principles and implemented using an infrastructure-as-code strategy, SPLICE enables reproducible deployment across local systems, national-scale facilities, commercial cloud platforms and trusted research environments, ensuring flexibility, security and sustainability.

SPLICE is being adopted to meet the data management and analysis needs of CRC-STARS, the MRC National Mouse Genetics Network, and PREDICT-Meso projects. Its design is driven by a commitment to make data integration and collaboration simpler, more robust and more widely accessible. Key features include: automated analysis and reporting pipelines, data collection and curation aligned with FAIR best practices, support for sensitive data within trusted research environments, advanced search and visualisation capabilities for building custom datasets, and support for generating high-quality training datasets for AI applications.

We report on the first deployment of the platform at the Edinburgh International Data Facility, which is supporting the development of backend components and programmatic interfaces. The platform architecture includes four modules: data transfer and quality control, data storage and access, data analysis and visualisation, and data export. Together, these components form a coherent framework that supports high-throughput data movement, scalable analysis and streamlined research workflows.

Glioblastoma is an aggressive brain tumour with poor outcome. Studying gene expression aids our understanding of the biology of glioblastoma and identifies potential therapeutic targets. The RNA cap is added to RNA polymerase II-transcribed RNAs and has important roles in regulating gene expression. The RNA cap methyltransferase CMTR1 has significant impacts in RNA translation, mitochondrial function and DNA replication. We targeted this enzyme in neural stem cells with glioblastoma-associated mutations and patient-derived glioblastoma cell lines and investigated its impacts on transcriptome. RNA-seq showed that many neural-specific transcripts were responsive to CMTR1 inhibition, which were developmental-related and/or cancer-associated. Moreover, we integrated the published bulk RNA-seq and snRNA-seq datasets of patient glioblastoma tumours into our analysis. We found that CMTR1 inhibition more specifically regulated mesenchymal (MES)-glioblastoma specific genes. Consistently, the tumours with increased proportion of MES subtype cells were associated higher CMTR1 expression. Furthermore, increased expression of CMTR1 also correlated with the proportion of cycling cells within the tumour, which implied the roles of CMTR1 in regulating the malignant cell proliferation. These observations suggested the gene/cell-type specificity of CMTR1 targets in glioblastoma tumours and highlighted the potentials of CMTR1 as a therapeutic target in glioblastoma.

High-risk neuroblastoma has a poor prognosis, owing to frequent relapse driven by acquired treatment resistance. Reversible switching between cell states, termed phenotypic plasticity, is increasingly recognised as a mechanism underlying treatment resistance in paediatric cancers. In high-risk neuroblastoma, tumour cells can switch between treatment-sensitive (adrenergic) and treatment-resistant (mesenchymal) phenotypes, and recent work suggests that plasticity traits are heritable and selected for during clonal evolution. Here, we investigate under what conditions phenotypic switching rates are identifiable from static and evolving barcode data.

We evaluate the identifiability of phenotypic switching rates in a population-level framework, combining structural analysis of analytical moment equations with Approximate Bayesian Computation. Experimental design strongly affects recovery: founding phenotype and observation time both influence inference, with longer observation times reducing information in the correlation structure of phenotype counts. However, recovery is constrained by the fitness cost of the resistant state, and slower switching rates governing rare subpopulations are often unrecoverable. Within the population-level ABC framework, variance in the proportion of adrenergic cells is the most informative statistic for inferring switching rates. We then investigate single-cell phylogeny-based statistics as a complementary source of information, finding that they exhibit structurally distinct features across the parameter space and therefore may improve inference beyond count-based summaries.

Osteoarthritis (OA) is a leading cause of chronic pain and disability and currently lacks effective disease-modifying therapies. Although cartilage degeneration is a hallmark of OA, increasing evidence supports a critical role for subchondral bone in disease progression and pain. Osteocytes, are the master regulators of bone homeostasis and may also influence bone–nerve interactions. However, their role in OA-related pain mechanisms remains poorly understood.

We hypothesized that genetically determined disruption of key pathways leads to abnormal osteocyte function and nociceptive neuronal sprouting in bone. We therefore focused on LTBP3, an OA gene and regulator of TGF-β signalling.

We generated an ltbp3 zebrafish mutant, and used micro-computed tomography and Raman confocal microscopy for skeletal phenotyping. Adult ltbp3 mutants showed an abnormal skeletal phenotype characterised by scoliosis, sclerosis, along with a reduced number and altered morphology of osteocytes. To investigate Ltbp3 in a mammalian system, we generated Ltbp3 KO MLO-Y4 osteocytes and characterised their morphology using Cell Painting, which showed abnormal elongation and enhanced dendritic branching. In parallel, pSMAD2/3 immunostaining revealed reduced activation of the TGF-β pathway, consistent with the known role of LTBP3 in regulating TGF-β signalling. To explore whether Ltbp3 changes in osteocyte, RNA-seq and secretome analysis were performed, revealing differential expression of genes and proteins involved in cytoskeletal organisation, extracellular matrix regulation, and interestingly axon guidance. To study whether Ltbp3 KO MLO-Y4 secreted proteins influence sensory neurons, we cultured immortalised Dorsal Root Ganglia with conditioned media. Remarkably, we observed an increased neuronal migration, supporting the role of osteocytes in a pro-innervation neuro-skeletal environment. Next, we analysed nociceptor neurons in joint of ltbp3 zebrafish mutant. We detected disorganised sensory sprouting, alongside direct contact between neurons and bone cells.

Our findings identify LTBP3 as a mediator of neuro-skeletal cross talk and possible pain modulator in OA.