Explore the vast range of titles available.

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

Metastasis is the major cause of cancer-related deaths. Neuroblastoma (NB), a childhood tumor has been molecularly defined at the primary cancer site, however, the bone marrow (BM) as the metastatic niche of NB is poorly characterized. Here we perform single-cell transcriptomic and epigenomic profiling of BM aspirates from 11 subjects spanning three major NB subtypes and compare these to five age-matched and metastasis-free BM, followed by in-depth single cell analyses of tissue diversity and cell-cell interactions, as well as functional validation. We show that cellular plasticity of NB tumor cells is conserved upon metastasis and tumor cell type composition is NB subtype-dependent. NB cells signal to the BM microenvironment, rewiring via macrophage mgration inhibitory factor and midkine signaling specifically monocytes, which exhibit M1 and M2 features, are marked by activation of pro- and anti-inflammatory programs, and express tumor-promoting factors, reminiscent of tumor-associated macrophages. The interactions and pathways characterized in our study provide the basis for therapeutic approaches that target tumor-to-microenvironment interactions. The bone marrow is a common site of metastasis for neuroblastoma patients. Here, the authors perform single cell RNA-seq and ATAC-seq of bone marrow aspirates from 16 subjects and show conservation of tumor cell plasticity in metastases and identify tumor-to-bone marrow cell signals that trigger tumor promoting monocytes.

Sleep in mammals can be broadly classified into two different physiological categories: rapid eye movement (REM) sleep and slow-wave sleep (SWS), and accordingly REM and SWS are thought to achieve a different set of functions. The fruit fly Drosophila melanogaster is increasingly being used as a model to understand sleep functions, although it remains unclear if the fly brain also engages in different kinds of sleep as well. Here, we compare two commonly used approaches for studying sleep experimentally in Drosophila : optogenetic activation of sleep-promoting neurons and provision of a sleep-promoting drug, gaboxadol. We find that these different sleep-induction methods have similar effects on increasing sleep duration, but divergent effects on brain activity. Transcriptomic analysis reveals that drug-induced deep sleep (‘quiet’ sleep) mostly downregulates metabolism genes, whereas optogenetic ‘active’ sleep upregulates a wide range of genes relevant to normal waking functions. This suggests that optogenetics and pharmacological induction of sleep in Drosophila promote different features of sleep, which engage different sets of genes to achieve their respective functions.

Emerging data demonstrate homologous recombination (HR) defects in castration-resistant prostate cancers, rendering these tumours sensitive to PARP inhibition. Here we demonstrate a direct requirement for the androgen receptor (AR) to maintain HR gene expression and HR activity in prostate cancer. We show that PARP-mediated repair pathways are upregulated in prostate cancer following androgen-deprivation therapy (ADT). Furthermore, upregulation of PARP activity is essential for the survival of prostate cancer cells and we demonstrate a synthetic lethality between ADT and PARP inhibition in vivo. Our data suggest that ADT can functionally impair HR prior to the development of castration resistance and that, this potentially could be exploited therapeutically using PARP inhibitors in combination with androgen-deprivation therapy upfront in advanced or high-risk prostate cancer. Tumours with homologous recombination (HR) defects become sensitive to PARPi. Here, the authors show that androgen receptor (AR) regulates HR and AR inhibition activates the PARP pathway in vivo, thus inhibition of both AR and PARP is required for effective treatment of high risk prostate cancer.

BackgroundIntraoperative radiation therapy (IORT) has been investigated for patients with low-risk, early-stage breast cancer. The The North American experience was evaluated by TARGIT-R (retrospective) to provide outcomes for patients treated in “real-world” clinical practice with breast IORT. This analysis presents a 5-year follow-up assessment.MethodsTARGIT-R is a multi-institutional retrospective registry of patients who underwent lumpectomy and IORT between the years 2007 and 2013. The primary outcome of the evaluation was ipsilateral breast tumor recurrence (IBTR).ResultsThe evaluation included 667 patients with a median follow-up period of 5.1 years. Primary IORT (IORT at the time of lumpectomy) was performed for 72%, delayed IORT (after lumpectomy) for 3%, intended boost for 8%, and unintended boost (primary IORT followed by whole-breast radiation) for 17% of the patients. At 5 years, IBTR was 6.6% for all the patients, with 8% for the primary IORT cohort and 1.7% for the unintended-boost cohort. No recurrences were identified in the delayed IORT or intended-boost cohorts. Noncompliance with endocrine therapy (ET) was associated with higher IBTR risk (hazard ratio [HR], 3.67). Patients treated with primary IORT who were complaint with ET had a 5-year IBTR rate of 3.9%.ConclusionThe local recurrence rates in this series differ slightly from recent results of randomized IORT trials and are notably higher than in previous published studies using whole-breast radiotherapy for similar patients with early-stage breast cancer. Understanding differences in this retrospective series and the prospective trials will be critical to optimizing patient selection and outcomes going forward.

We identify biomarkers for disease progression in three type 2 diabetes cohorts encompassing 2,973 individuals across three molecular classes, metabolites, lipids and proteins. Homocitrulline, isoleucine and 2-aminoadipic acid, eight triacylglycerol species, and lowered sphingomyelin 42:2;2 levels are predictive of faster progression towards insulin requirement. Of ~1,300 proteins examined in two cohorts, levels of GDF15/MIC-1, IL-18Ra, CRELD1, NogoR, FAS, and ENPP7 are associated with faster progression, whilst SMAC/DIABLO, SPOCK1 and HEMK2 predict lower progression rates. In an external replication, proteins and lipids are associated with diabetes incidence and prevalence. NogoR/RTN4R injection improved glucose tolerance in high fat-fed male mice but impaired it in male db/db mice. High NogoR levels led to islet cell apoptosis, and IL-18R antagonised inflammatory IL-18 signalling towards nuclear factor kappa-B in vitro. This comprehensive, multi-disciplinary approach thus identifies biomarkers with potential prognostic utility, provides evidence for possible disease mechanisms, and identifies potential therapeutic avenues to slow diabetes progression. There is an urgent need for biomarkers for type 2 diabetes progression that provide a deeper understanding of the disease process. Here, the authors identify biomarkers in three molecular classes, replicate them in other cohorts and explore top protein biomarkers in detail in functional studies.

The reported influence of donor Killer-cell Immunoglobulin-like Receptor (KIR) genes on the outcomes of haematopoietic cell transplantation (HCT) are contradictory, in part due to diversity of disease, donor sources, era and conditioning regimens within and between different studies. Here, we describe the results of a retrospective clinical analysis establishing the effect of donor KIR motifs on the outcomes of 119 HLA-matched, unrelated donor HCT for adult acute myeloid leukaemia (AML) using myeloablative conditioning (MAC) in a predominantly T-cell deplete (TCD) cohort. We observed that HCT involving donors with at least one KIR B haplotype were more likely to result in non-relapse mortality (NRM) than HCT involving donors with two KIR A haplotypes (p = 0.019). Upon separation of KIR haplotypes into their centromeric (Cen) and telomeric (Tel) motif structures, we demonstrated that the Cen-B motif was largely responsible for this effect (p = 0.001). When the cause of NRM was investigated further, infection was the dominant cause of death (p = 0.006). No evidence correlating donor KIR B haplotype with relapse risk was observed. The results from this analysis confirm previous findings in the unrelated, TCD, MAC transplant setting and imply a protective role for donor-encoded Cen-A motifs against infection in allogeneic HCT recipients.

Background Depression is a common mental health condition now viewed as chronic or long-term. More than 50 % of people will have at least one further episode of depression after their first, and therefore it requires long-term management. However, little is known about the effectiveness of self-management in depression, in particular from the patients’ perspective. This study aimed to understand how people with longer-term depression manage the condition, how services can best support self-management and whether the principles and concepts of the recovery approach would be advantageous. Methods Semi-structured in depth interviews were carried out with 21 participants, recruited from a range of sources using maximum variation sampling. Interpretative Phenomenological Analysis was used by a diverse team comprised of service users, practitioners and academics. Results Four super-ordinate themes were found: experience of depression, the self, the wider environment, self-management strategies. Within these, several prominent sub-themes emerged of importance to the participants. These included how aspects of themselves such as hope, confidence and motivation could be powerful agents; and how engaging in a wide range of chosen activities could contribute to their emotional, mental, physical, social, spiritual and creative wellbeing. Conclusions Services in general were not perceived to be useful in specifically facilitating self-management. Increased choice and control were needed and a greater emphasis on an individualised holistic model. Improved information was needed about how to develop strategies and locate resources, especially during the first episode of depression. These concepts echoed those of the recovery approach, which could therefore be seen as valuable in aiding the self-management of depression.