Childhood T-cell acute lymphoblastic leukaemia (T-ALL) is a malignancy of developing T cells and carries a worse clinical prognosis than its B-cell counterpart (B-ALL). The origin of childhood T-ALL that does not respond to initial treatment, i.e. refractory disease, is unknown. Refractory T-ALL carries a poor prognosis and cannot be predicted at diagnosis. Here, we performed single-cell RNA sequencing (scRNA-seq) of T-ALL from children who did or did not respond to initial treatment. We identified a transcriptionally distinctive blast population, exhibiting features of innate-like lymphocytes, as the major source of refractory disease. Evidence of such blasts at diagnosis heralded refractory disease across independent datasets and was associated with survival in a large, contemporary trial cohort. Moreover, for one individual, by projecting somatic mutations onto scRNA-seq data, we demonstrated that these non-canonical blasts were detectable at diagnosis as a minor subclone (~1% of blasts), but expanded dramatically following treatment to become the dominant clone. Our findings portray refractory T-ALL as a distinct disease driven by a previously unknown lymphoblast variant and may have immediate clinical utility.

Diffuse large B-cell lymphoma (DLBCL) is the commonest form of non-Hodgkin lymphoma. Recent studies have revealed its considerable genetic heterogeneity and elucidated diverse oncogenic pathways that drive distinct molecular subtypes of DLBCL, highlighting the potential for precision medicine approaches. However, a barrier to implementing patient-personalised, targeted therapies in DLBCL is that existing assays for real-time genetic profiling, as well as those to monitor response to therapy, suffer from logistical and technical limitations. Circulating tumour DNA (ctDNA) has potential to overcome these hurdles. However, there is widespread technical variation between different ctDNA assays reported across recent studies. Our study highlights the feasibility and logistical advantages of plasma-based genotyping when assay turnaround time is considered in the context of a multisite study. However, we also highlight limitations of plasma-based genotyping for those patients with very low tumour volume, and define clinical predictors of genotyping success. We compare single nucleotide variant (SNV) with copy number-based quantification of pretreatment ctDNA burden. Although both approaches correlate equally well with radiological assessment of tumour burden, copy number-based assessment identifies an especially poor risk cohort of patients with high tumour burden that may be prioritised for experimental therapy in future trials. Finally, we optimise and assess the performance of a phased variant-supported, noise suppressed, ultrasensitive quantification of milimal residual disease (MRD). We describe the determinants of assay sensitivity and their impact on limit of detection at a per-patient level. We show how the achievement of MRD negativity is an extremely strong predictor of lymphoma-free survival.

With ageing, normal tissues become colonised by expanding clones, which acquired fitness-enhancing mutations that allow them to over-proliferate relative to their neighbours. However, whether the clones’ founding cell also carries an epigenome that is inherited by the expanding clone is incompletely understood. In the following study, we delineate the fundamental properties of how epigenetic states are inherited across the lifespan in human hematopoietic stem cells (HSCs). We performed single-cell whole genome and methylation sequencing of HSCs from several healthy and diseased individuals. We built phylogenies of HSCs and combined them with read counts at CpG sites to accurately infer methylation state changes in millions of CpG sites. Using our bespoke method, we were able to determine the origin of each of these methylation changes and map them onto the phylogeny. Our method fully recapitulates the natural dynamics of methylation and allows us to accurately time methylation changes to the human lifespan. We analysed approximately 25 million CpG sites per individual, and we find most sites exhibit heritability of methylation states. We found the rate of methylation change is several folds higher than the rate of somatic mutations and that methylation changes are stochastic and allele-specific. Methylation changes are remarkably stable as we found that hundreds of changes are acquired between the time of X-chromosome inactivation and germ layer formation and are stably inherited until old age. In addition, we find that some of these changes precede the neoplasm’s clone, which suggests they might play a role in function.

The Leap ΔTissue Program has generated high volumes of clinical metadata and experimental data in various formats using diverse modalities to study three research areas: Glioblastoma, Triple Negative Breast Cancer and Tuberculosis. Efficient access, organisation and integration of these are essential for accelerating discovery, linking and re-use of data generated within the program. We outline the case study of using glioblastoma data generated within the program, to set-out a linked data pipeline that facilitates querying between different projects and teams within the program. We aim to develop a Linked Data Resource, which will function similarly to Wikidata but will incorporate data generated within the program. This resource will allow for sophisticated querying of both experimental and clinical metadata. We applied the FAIR data management principles (Findable, Accessible, Interoperable, and Reusable) to curate and link Glioblastoma (CNS WHO Grade 4) experimental and clinical metadata between two groups: the Pages Group, who are exploring Glioblastoma microenvironment in 3D and the Bayraktar Group; who are mapping the tissue architecture of brain tumours through the integration of single cell and spatial transcriptomics. In the Linked Data Glioblastoma project we identified broader data structures that describe the experimental and clinical metadata from these two groups. We standardised entries by assigning them unique identifiers from ontologies or community standards and making them machine-readable. The data formats were then unified into RDF (Resource Description Framework). These unified data structures and pipelines can be reused across other projects, streamlining data sharing, data integration and querying processes. Linked Data enables efficient data integration, reuse, and advanced querying of experimental and clinical metadata. The linked data glioblastoma project serves as a blueprint for scaling this pipeline to other research areas within the ΔTissue program and can act as guidance for the creation of a centralised department-wide linked data resource.

Background: In 2011, Kenya introduced PCV10 as a preventive measure for invasive pneumococcal disease (IPD). However, sustained disease control is compromised by the emergence of disease caused by non-vaccine serotypes (nVT). Methods: We evaluated the impact of PCV10 in Kenya using whole-genome sequencing, by carrying out a cross-sectional analysis of 3080 isolates obtained from healthy carriers and patients with IPD from Kilifi, Kenya pre- and post-PCV10 (GSK). We compared the distribution of serotypes, global pneumococcal sequence clusters (GPSCs), and antimicrobial resistance (AMR) pre- and post-PCV10 era. Results: We report a diverse pneumococcal population consisting of 58 serotypes and 103 GPSCs. The most predominant serotype in this population was serotype 1 (7.2%) belonging to [GPSC2], which had the highest proportion in the disease group before and after vaccine introduction, while in the carriage group, serotypes 19F and 19A had the highest proportion at (16% and 7%), respectively. GPSC22 was the most dominant lineage at 8.7% and consisted of nine serotypes with serotype 15A leading with 46.4%. In the disease group, GPSC2 dominated pre-and post-PCV10 with 30% and 15% respectively, while GPSC 879 and GPSC22 dominated in the carriage group with 11% and 13%, respectively. In the overall population, there was a significant reduction in VTs and an increase in NVTs which was consistent in both disease and carriage groups. The most increased lineages post-PCV10 was GPSC891 [OR=24.9] in disease and GPSC883 [OR=11.6] in carriage group. AMR genes associated with penicillin, cotrimoxazole, tetracycline and doxycycline increased with PCV10 introduction, and in lineages that emerged after vaccine introduction. Conclusion: The study reported the PCV10 impact on the population structure in Kilifi Kenya.

Endoscopes typically use red, green and blue to mimic human vision in hard-to-reach places... but human vision isn't very good at seeing cancer. I am developing targeted multispectral endoscopy systems to improve detection of cancer and precancerous diseases. I will outline the design and fabrication of our custom snapshot multispectral sensors, optimised for simultaneous imaging of blood vasculature and oxygenation, and surgical dye fluorescence alongside familiar RGB images for navigation. I will also introduce our clinical trial, evaluating the efficacy of spectral imaging for delineating cancer, precancerous lesions and benign tissue in the common disease "Barrett's Oesophagus". Barrett's currently suffers from a 50% miss rate of early precancerous changes with traditional RGB cameras; improved detection is essential for effective triage and treatment.

Nipah virus (NiV), a highly lethal virus in humans, circulates in Pteropus bats throughout South and Southeast Asia. Difficulty in obtaining viral genomes from bats means we have a poor understanding of NiV diversity. We develop phylogenetic approaches applied to the most comprehensive collection of genomes to date (N=257, 175 from bats, 73 from humans) from six countries over 22 years (1999-2020). We divide the four major NiV sublineages into 15 genetic clusters. Using Approximate Bayesian Computation fit to a spatial signature of viral diversity, we estimate the presence and the average size of genetic clusters per area. We find that, within any bat roost, there are an average of 2.4 co-circulating genetic clusters, rising to 5.5 clusters at areas of 1500-2000km2. We estimate that each genetic cluster occupies an average area of 1.3million km2 (95%CI: 0.6-2.3 million), with 14 clusters in an area of 100,000km2 (95%CI: 6-24). In the few sites in Bangladesh and Cambodia where genomic surveillance has been concentrated, we estimate that most clusters have been identified, but only ~15% of overall NiV diversity has been uncovered. Our findings are consistent with entrenched co-circulation of distinct lineages, even within roosts, coupled with slow migration over larger spatial scales.

Activin A is a TGF-β family ligand involved in many biological processes related to growth and development. As a result, activin A signalling also contributes to the progression of various cancers, in which it has been found to promote cell proliferation, stem-like traits, metastasis, and primary tumour development. There is a wide interest in the development of novel activin A inhibitors to treat cancer. We have discovered a monoclonal antibody that binds activin A and closely related ligand activin B with low nanomolar affinity. The antibody can neutralise bioactivity of activins and competes for the type II receptor ectodomain. We present characterisation of the antibody binding to activins using biophysical methods. We have also used cryo-EM to determine low-resolution structure of the full complex with binding stoichiometry of 2 FAb:1 activin A dimer. While the epitope could not be mapped at the atomic level, this structure supports the expected mechanism of action of the antibody as blocking type II receptor binding at a conserved epitope between activins A and B.

Over 700,000 DNA sequences can form non-canonical DNA structures - G-quadruplexes (G4s). However, only a small subset of these motifs forms structures in cells. These G4 landscapes are cell type-specific and enriched in enhancers. This could suggest that G-quadruplexes could play a role in determining cell state through gene expression regulation. I will present our work on studying the role of G-quadruplexes in these cis-regulatory elements, using the alpha and beta globin locus control regions as examples. Several models have been proposed for exactly how enhancers regulate gene expression at a linear distance; one prominent model involves interactions in 3D space between the enhancer and gene. We have dissected the role of G4s in such interactions as well as characterising the effect of abolishing and restoring the structure on enhancer function. We have also made inroads into identifying factors that could explain these functional effects.

Genetic variants linked to autism are thought to change cognition and behaviour by altering the structure and function of the brain. Although a substantial body of literature has identified structural brain differences in autism, it is unknown whether autism-associated common genetic variants are linked to changes in cortical macro- and micro-structure. We investigated this using neuroimaging and genetic data from adults (UK Biobank, N = 31,748) and children (ABCD, N = 4,928). Using polygenic scores and genetic correlations we observe a robust negative association between common variants for autism and a magnetic resonance imaging derived phenotype for neurite density (intracellular volume fraction) in the general population. This result is consistent across both children and adults, in both the cortex and in white matter tracts, and confirmed using polygenic scores and genetic correlations. There were no sex differences in this association. Mendelian randomisation analyses provide no evidence for a causal relationship between autism and intracellular volume fraction, although this should be revisited using better powered instruments. Overall, this study provides evidence for shared common variant genetics between autism and cortical neurite density.