Biology subjects

Marsden, B.

Publications and source records attributed to Marsden, B..

6 recordsLinked to original sources

bioRxiv2026

Distinct transcriptional states of draining lymph node T cells following influenza vaccination in individuals with underrepresented ancestries

The transcriptional and cell repertoire changes of lymph node (LN) T cells following influenza vaccination have been under-investigated across different ethnicities. Draining (dLN) and non-draining (ndLN) axillary LN cells from 17 participants with genotypically varied African or Asian ancestry (ISRCTN13657999) were collected by fine needle aspiration (FNA) before, 5 days and 42 days after adjuvanted influenza vaccination and analysed by single-cell RNA-sequencing (scRNA-seq). The early dLN transcriptional profile and cellular dynamics were distinct from the ndLN; BACH2 expressing T cells were highly induced, exhibiting stem-like and unique quiescence signatures. Proliferating T follicular helper cells (Tfh) were enriched in dLN and Th1-like Tfh predominated with highly induced transcriptional changes. The BACH2+ T cell state and Th1-like Tfh cells, transcriptionally distinct from germinal centre (GC) T cells, were co-localised with LAMP3+ dendritic cells in the T cell zone of publicly available LN datasets, a region that was also distinct from GC Tfh niches. Together, these findings indicate that the induction of a robust, proliferative, and transcriptionally diverse Tfh response, alongside a discrete subset of BACH2+ T cells, represents a marked feature of the early T cell zone immune response to adjuvanted influenza vaccine in the dLN.

immunology↗

bioRxiv2025

INTERPOLATION-BASED CONDITIONING OF FLOW MATCHING MODELS FOR BIOISOSTERIC LIGAND DESIGN

AO_SCPLOWBSTRACTC_SCPLOWFast, unconditional 3D generative models can now produce high-quality molecules, but adapting them for specific design tasks often requires costly retraining. To address this, we introduce two training-free, inference-time conditioning strategies, Interpolate-Integrate and Replacement Guidance, that provide control over E(3)-equivariant flow-matching models. Our methods generate bioisosteric 3D molecules by conditioning on seed ligands or fragment sets to preserve key determinants like shape and pharmacophore patterns, without requiring the original fragment atoms to be present. We demonstrate their effectiveness on three drug-relevant tasks: natural product ligand hopping, bioisosteric fragment merging, and pharmacophore merging.

molecular biology↗

bioRxiv2024

MolSnapper: Conditioning Diffusion for Structure Based Drug Design

Generative models have emerged as potentially powerful methods for molecular design, yet challenges persist in generating molecules that effectively bind to the intended target. The ability to control the design process and incorporate prior knowledge would be highly beneficial for better tailoring molecules to fit specific binding sites. In this paper, we introduce MolSnapper, a novel tool that is able to condition diffusion models for structure-based drug design by seamlessly integrating expert knowledge in the form of 3D pharmacophores. We demonstrate through comprehensive testing on both the CrossDocked and Binding MOAD datasets, that our method generates molecules better tailored to fit a given binding site, achieving high structural and chemical similarity to the original molecules. It also, when compared to alternative methods, yields approximately twice as many valid molecules.

bioinformatics↗

bioRxiv2023

Robustly interrogating machine learning based scoring functions: what are they learning?

MotivationMachine learning-based scoring functions (MLBSFs) have been found to exhibit inconsistent performance on different benchmarks and be prone to learning dataset bias. For the field to develop MLBSFs that learn a generalisable understanding of physics, a more rigorous understanding of how they perform is required. ResultsIn this work, we compared the performance of a diverse set of popular MLBSFs (RFScore, SIGN, OnionNet-2, Pafnucy, and PointVS) to our proposed baseline models that can only learn dataset biases on a range of benchmarks. We found that these baseline models were competitive in accuracy to these MLBSFs in almost all proposed benchmarks, indicating these models only learn dataset biases. Our tests and provided platform, ToolBoxSF, will enable researchers to robustly interrogate MLBSF performance and determine the effect of dataset biases on their predictions. Availability and Implementationhttps://github.com/guydurant/toolboxsf Contactdeane@stats.ox.ac.uk Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2023

Pericardial & Mediastinal Fat-Associated Lymphoid Clusters are rapidly activated in an alkane induced model of Systemic Lupus Erythematosus

Systemic Lupus Erythematosus (SLE) is an autoimmune disease predominated by auto-antibodies that recognise cellular components. Pleural involvement is the most common SLE-related lung disease. Natural antibodies are rapidly secreted by innate-like B cells following perturbation of homeostasis and are important in the early stages of immune activation. The serous cavities are home to large numbers of innate-like B cells present both within serous fluid and resident within fat-associated lymphoid clusters (FALCs). FALCs are important hubs for B-cell activation and local antibody secretion within the body cavities. Patients with SLE can develop anti-phospholipid antibodies and in rare situations develop alveolar haemorrhage. Utilising delivery of the hydrocarbon oil pristane in C57BL/6 mice as a model of SLE we identify a rapid expansion of pleural cavity B cells as early as day 3 after intra-peritoneal pristane delivery. Following pristane delivery, pericardial B1 B cells are proliferative, express the plasma-cell surface marker CD138 and secrete both innate and class switched antibodies highlighting that this cavity niche may play an unrecognised role in the initiation of lupus pleuritis. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=163 SRC="FIGDIR/small/549766v1_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@18ea5bforg.highwire.dtl.DTLVardef@23d9deorg.highwire.dtl.DTLVardef@1b48b3aorg.highwire.dtl.DTLVardef@20a133_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2018

SATurn: A modular bioinformatics framework for the design of robust maintainable web-based and standalone applications

SummarySATurn is a modular, open-source, bioinformatics platform designed to specifically address the problems of maintenance and longevity commonly associated with the development of simple tools funded by academic research grants. Applications developed in SATurn can be deployed as web-based tools, standalone applications, or hybrid tools which have the benefits of both. Within the Structural Genomics Consortium (SGC) we have utilized SATurn to create a bioinformatics portal which routinely supports a diverse group of scientists including those interested in structural biology, cloning, glycobiology & chemicalbiology.\n\nSupplementary information:

bioinformatics↗