University of Exeter

Setting up a psychedelic study can be a long, arduous, and Kafkaesque process. Researchers are faced with a host of challenges in this rapidly evolving field, necessitating a range of considerations that remain largely unstandardised. Many of the complexities inherent to psychedelic research also challenge existing assumptions around, for example, approaches to psychiatric prescribing, the conceptual framing of the placebo effect, and definitions of selfhood. This review paper aims to formalise these unique considerations by addressing the sociocultural, political, legal, pharmacological, safety, study design and experiential facets inherent to a psychedelic study. We bring together several of the major psychedelic research teams across the United Kingdom, identify continuing areas of debate, and provide a practical, comprehensive, experience-based guide, with recommendations for policymakers and future researchers intending to set up a psychedelic research study or clinical trial.

There is evidence that over one quarter of patients return to National Health Service (NHS) talking therapies (TT) services, and the needs of these patients are poorly understood and catered for. This project investigated the acceptability and effectiveness of delivering cognitive analytic therapy (CAT) for patients with depression returning to a TT service with childhood trauma and associated relational difficulties. A case-controlled pilot study using TT sessional outcome measures. A 16-session CAT was offered to N = 76 patients who had previously received a high-intensity intervention (mainly cognitive-behavioural therapy; CBT) in the same TT service. Dropout rates, recovery rates and when recovery occurred during the CAT treatment episode were calculated. CAT outcomes were compared against the previous treatment episodes (n = 47) and also benchmarked against the evidence base. Patients were followed up after receiving CAT (n = 16) to assess the durability of change. The number then returning to the TT service after receiving CAT was also tracked. The dropout rate for CAT was 16.9% and the reliable recovery rate was 40%. Reliable and/or clinically significant reductions in depression tended to occur during early CAT sessions. At a group level, there were significant reductions in depression during CAT. There was no evidence of relapse at follow-up. The return rate to the service following CAT was 28.94%. CAT appears useful to consider in the offer for patients returning to TT services. Clinical trials now need to focus on the treatment return patient group in TT services across the stepped care model.

Understanding the evolution of pathogen host range is a challenging problem but one that is important for emerging infections and for biocontrol. Theory predicts that sequential selection in multiple host species should select for broad host range. Using two genotypes of B. thuringiensis we tested whether selection in alternating host environments would produce bacteria with increased virulence in both hosts, relative to bacteria passaged in single host species (fall armyworm, Spodoptera frugiperda or diamondback moth, Plutella xylostella). Since the Cry toxins of B. thuringiensis are public goods that benefit groups of bacteria, not individual cells, we employed a passage design which provided reproductive benefits to groups of pathogens, based on infectivity. Passage of one bacterial genotype (Bt morrisoni) led to the loss of virulence, while the second genotype (Bt galleriae) evolved virulence that was dependent on selection treatment. In contrast to expectation, selection in P. xylostella produced lineages with increases in virulence in both hosts; selection in S. frugiperda led to very low virulence and the alternating host treatment produced intermediate levels of virulence. Modest increases in virulence were accompanied by a reduction in fitness, consistent with a cost of increased investment in virulence factors. In contrast, infection in S. frugiperda selected for cheaters that had reduced investment in Cry toxins and high competitive fitness within hosts. In conclusion, the selection favouring cheaters depended strongly on both host species and bacterial genotype. Importantly, the host (P. xylostella) that favoured cooperation produced mutants with gains in virulence across multiple hosts.

The rapid expansion of data centers (DCs) has led to substantial increases in global energy consumption and carbon emissions. Moreover, the strong coupling among workload scheduling, IT, cooling, and energy subsystems makes sustainable DC microgrids highly complex, especially given the difficulty of accurate modeling and the presence of significant uncertainties and rapid dynamics. To address these challenges, this paper proposes a multi-critic deep reinforcement learning (DRL) approach based on a mixture-of-experts (MoE) with multi-head architecture. The problem is formulated as a multi-reward Markov Decision Process (MDP) with independent reward signals for workload efficiency, energy consumption, and carbon footprint, providing richer feedback compared to traditional single-reward formulations. In the proposed approach, each reward is assigned a dedicated critic, thereby avoiding the conflicts and interference that arise when a single critic attempts to learn multiple competing rewards simultaneously. The shared MoE foundation with the multi-head architecture enables each critic head to focus on learning the value function for its specific optimization target, while the integrated adaptive gating mechanisms facilitate dynamic leveraging of both common and task-specific knowledge. This architecture improves learning stability, accelerates convergence, and enhances adaptability to complex, multi-objective learning tasks. Extensive experiments on a DC microgrid show that the proposed method outperforms state-of-the-art DRL and rule-based baselines in reducing the overall energy consumption and carbon emissions. All codes can be found at https://github.com/ikelq/Mixture-of-experts-based-Multi-Critic-DRL-for-Sustainable-Management-of-Data-Center-Microgrids . • Formulate the sustainable management of a DC microgrid as a multi-reward MDP. • Propose a multi-critic DRL framework with each reward assigned to a dedicated critic. • Design a multi-head architecture with a shared MoE structure within the multi-critic DRL framework.

Non-protein-coding genes are emerging as critical contributors to the etiology of rare diseases, providing key insights into human biology and uncovering novel disease mechanisms. We identified 7 individuals from 4 families with early-onset diabetes (diagnosed aged <5 years) and immune dysregulatory features caused by bi-allelic variants in RNU6ATAC. RNU6ATAC encodes a small nuclear RNA (snRNA) that acts as a catalytic component of the minor spliceosome, a protein-RNA complex that mediates the splicing of ∼700 genes containing U12/minor-type introns. Variant screening of the other 64 minor spliceosome genes in 276 infants with diabetes identified 12 unrelated individuals with bi-allelic disease-causing variants in RNU4ATAC. Bi-allelic pathogenic RNU4ATAC variants are known to cause a variable spectrum of clinical features, which until now did not include diabetes. Clinically, 12/19 RNU6ATAC/RNU4ATAC affected individuals had additional immune dysregulatory features, and 50% of individuals tested were islet-autoantibody positive, strongly supporting an autoimmune etiology for their diabetes. RNA sequencing (RNA-seq) in 3 individuals with bi-allelic RNU6ATAC variants showed a pattern of intron retention in U12-intron-containing genes similar to that seen in RNU4ATAC individuals (n = 3), supporting a shared disease mechanism. Analysis of affected individuals' transcriptomic, methylation, and immune data revealed impaired B cell development and maturation. We conclude that bi-allelic RNU6ATAC variants cause a syndrome of early-onset autoimmune diabetes and immune dysregulation. We further show that infancy-onset diabetes is a feature of RNU4ATAC-opathy. Our work highlights the important role of two snRNAs critical to minor spliceosome function in immune system regulation, providing insights into the pathogenesis of autoimmune diabetes.

Direct current stimulation (DCS) -i.e., the application of a static electric field to the brain through surface scalp electrodes- emerged as a novel therapy for neuropsychiatric disorders. In this study, DCS-induced changes in gene expression were assessed on Botryllus schlosseri, a ubiquitous simple colonial chordate living in temperate seas and sharing genetic remarkable similarities with mammals. Colonies underwent either sham or real DCS stimulation. Behavioral assessments, including the Siphon Stimulation Test and heart-rate measurements, were conducted before and at three intervals post-stimulation (3, 24, and 48h). Gene expression was analyzed with sequencing reads via Trimmomatic®. Analysis of heart rate frequency revealed post-treatment differences between the groups immediately after stimulation (p=0.001), with the effect decreasing over time. Real DCS differentially expressed 191 genes at three h post-DCS, 104 at 24h, and 529 at 48h. Our study introduces a novel pre-clinical model for DCS application demonstrating for the first in vivo time an after-effect on gene expression related to inflammation, human development, and neurodegeneration. To the best of our knowledge, this is the first paper addressing DCS effects on gene expression in a living animal model.