Biology subjects

Zhou, Q.

Publications and source records attributed to Zhou, Q..

At least 91 records · Page 5Linked to original sources

bioRxiv2025

SpecImmune accurately genotypes diverse immune-related gene families using long-read data

Polymorphic immune-related genes (HLA, KIR, IG, TCR, and CYP) exhibit significant complexity due to their extensive heterozygosity and inter-loci homology, necessitating specific methods for accurate characterization. We present SpecImmune, the first comprehensive tool leveraging long-read sequencing data to resolve the full spectrum of these immune-related genes. The method adopts an iterative graph-based algorithm for haplotype reconstruction. We validated SpecImmune across 1,019 samples from the 1kGP ONT cohort, 42 PacBio CLR and 9 PacBio HiFi samples from the HGSVC project, and 47 PacBio HiFi and 37 ONT samples from the HPRC project. SpecImmune achieved an accuracy of 98% in HLA typing, which represents a 12% improvement over both SpecHLA and HLA*LA. SpecImmune is the initial method to type multiple CYP loci, as well as the foremost approach to allow precise KIR and germline IG/TCR typing using long reads. Comprehensive genotyping of these loci by SpecImmune unveils a new observation of substantial linkage disequilibrium among HLA, KIR, and CYP loci. The proteins derived from these loci exhibit strong binding affinities, which suggest the origin of the marked linkage disequilibrium. Further, SpecImmune unveils a novel finding of elevated IG/TCR heterozygosity in African populations. Additionally, SpecImmune facilitates the detection of de novo mutations and enables allele-specific drug recommendations.

bioinformatics↗

bioRxiv2025

N-Power AI: A Specialized Agent Framework for Automated Sample Size and Power Analysis in Clinical Trial Design

BackgroundSample size and power analysis are essential in biomedical research and investigations, particularly in clinical trial design, as they ensure sufficient statistical power to detect meaningful effects. However, the complexity of these calculations often requires specialized statistical expertise, making the process inconvenient and limiting accessibility for researchers during early-stage study planning. MethodsWe developed N-Power AI, an agentic framework leveraging large language models (LLMs) to perform sample size and power calculations across diverse study designs. The framework consists of three specialized agents: the Function Agent, a perception and reasoning module that identifies appropriate statistical tests and corresponding R functions; the Calculation Agent, an action module that extracts parameters and executes precise computations; and the Reporting Agent, a presentation module that generates comprehensive, downloadable reports. N-Power AI and advanced LLMs (e.g., GPT-o1, Claude 3.5, Gemini 1.5 Pro) were evaluated against ground truths from statistical software (R) across six common clinical trial scenarios. ResultsDirect LLM outputs showed significant deviations from ground-truth values, particularly in complex scenarios like the Chi-Square Test and Cox Proportional Hazards Model. In contrast, N-Power AI achieved 100% agreement with ground truths across all scenarios. This accuracy is attributed to the Function Agents correct selection of statistical methods, the Calculation Agents accurate computations, and the Reporting Agents ability to produce clear and comprehensive summaries. ConclusionN-Power AI automates sample size and power analysis, offering an effective, efficient, and accessible solution for early-stage study planning. While human expertise remains crucial for high-level statistical planning, N-Power AI enhances accessibility and efficiency, streamlining the analysis process to generate reliable and reproducible results for a wide range of research scenarios.

bioinformatics↗

bioRxiv2025

Niche-driven phenotypic plasticity and cis-regulatory dynamics of a revised model for intestinal secretory differentiation

HIGHLIGHTSO_LIDelineation of chromatin and mRNA dynamics of intestinal secretory differentiation C_LIO_LIPaneth cells have few unique enhancers and share mRNAs and TFs with goblet cells C_LIO_LIUnlike other secretory derivatives, goblet and Paneth cells are not specified per se C_LIO_LINiche factors, especially BMP signaling, define goblet and Paneth phenotypes C_LI Enterocytes and four secretory cell types derive from stem cells located in intestinal crypts. Whereas secretory goblet and Paneth cells have long been considered distinct, we find high overlap in their transcripts and sites of accessible chromatin, in marked contrast to those of sibling enteroendocrine or tuft cells. Mouse and human goblet and Paneth cells express extraordinary fractions of selective antimicrobial genes, reflecting specific and variable gene responses to local niche signals. Wnt signaling retains few ATOH1+ secretory daughters in crypt bottoms, where an absence of BMP signaling potently induces Paneth features; those that move away from crypt bottoms acquire classic goblet properties. These post-mitotic cellular phenotypes and their underlying accessible cis-elements interconvert readily. Thus, goblet and Paneth properties represent alternative manifestations of a single versatile signal-responsive secretory cell. These findings reveal exquisite niche-dependent cell plasticity and the cis-regulatory dynamics of an updated unitarian model of the intestinal epithelial lineage.

cell biology↗

bioRxiv2025

WDR2 regulates the orphan kinesin KIN-G to promote hook complex and Golgi biogenesis in Trypanosoma brucei

The flagellum in Trypanosoma brucei plays crucial roles in cell locomotion, cell morphogenesis, and cell division, and its inheritance depends on the faithful duplication of multiple flagellum-associated structures. One of such cytoskeletal structures is a hairpin-like structure termed the hook complex composed of a fishhook-like structure and a centrin arm structure, whose cellular functions remain poorly understood. We recently identified KIN-G, an orphan kinesin that promotes hook complex and Golgi biogenesis. Here we report a WD40 repeats-containing protein named WDR2, which interacts with and regulates KIN-G. WDR2 co-localizes with KIN-G at the centrin arm, and knockdown of WDR2 disrupts hook complex integrity and morphology, inhibits flagellum attachment zone elongation and flagellum positioning, and eventually arrests cytokinesis. Knockdown of WDR2 also disrupts the maturation of the centrin arm-associated Golgi, thereby impairing Golgi biogenesis. WDR2 interacts with KIN-G via its N-terminal unknown motifs, the middle domain containing a coiled coil and a PB1 motif, and the C-terminal WD40 domain, and targets KIN-G to its subcellular location. These results uncover a regulatory role for WDR2 in recruiting KIN-G to regulate hook complex and Golgi biogenesis, thereby impacting flagellum inheritance and cell division plane positioning.

cell biology↗

bioRxiv2025

Benchmarking and optimizing Perturb-seq in differentiating human pluripotent stem cells

Perturb-seq is a powerful approach to systematically assess how genes and enhancers impact the molecular and cellular pathways of development and disease. However, technical challenges have limited its application in stem cell-based systems. Here, we benchmarked Perturb-seq across multiple CRISPRi modalities, on diverse genomic targets, in multiple human pluripotent stem cells, during directed differentiation to multiple lineages, and across multiple sgRNA delivery systems. To ensure cost-effective production of large-scale Perturb-seq datasets as part of the Impact of Genomic Variants on Function (IGVF) consortium, our optimized protocol dynamically assesses experiment quality across the weeks-long procedure. Our analysis of 1,996,260 sequenced cells across benchmarking datasets reveals shared regulatory networks linking disease-associated enhancers and genes with downstream targets during cardiomyocyte differentiation. This study establishes open tools and resources for interrogating genome function during stem cell differentiation.

genomics↗

bioRxiv2025

Does sample size of leaf osmotic potential affect its relationship with cotton yield?

Leaf osmotic potential at full turgor ({pi}0) has been used frequently to indicate turgor loss point of plant leaves. However, even a rapid measurement of{pi} 0 using osmometry is time-consuming, if numerous leaf samples need to be measured. Because of this, researchers tend to use a small sample size to determine{pi} 0 and relate it to indices of crop performance. Yet the statistical and agronomic significance of using a small sample size of{pi} 0 to indicate crop performance is not known. We address this question using field measurements and statistical resampling. Six mature leaf samples were collected at the peak bloom stage from each of the 54 cotton plots in Texas, USA in 2024. The{pi} 0 of the collected leaves were measured using an osmometer. Seed cotton yields from the field plots were measured near the end of cotton season. To test the effect of sample size on strength of the linear relation between{pi} 0 and cotton yield, 1-6 resamples of{pi} 0 were randomly drawn with replacement from the original 6 measurements per plot for the 54 plots. The resampled data of{pi} 0 were then used as independent variable to predict cotton yield. We found that, considering the labor and cost, sampling 3 or 6 leaves per plot may not make a significant difference for the linear regression between{pi} 0 and cotton yield.

plant biology↗

bioRxiv2025

Molecular Insights into Cross-Species Spillover of Coronavirus HKU5 via ACE2 Receptor Recognition

Coronaviruses represent a significant zoonotic threat, with host adaptation serving as a pivotal determinant of cross-species transmission. The bat-derived {beta}-coronavirus HKU5 utilizes its spike (S) protein for receptor recognition and viral entry. Here, we report the cryo-electron microscopy (cryo-EM) structure of the HKU5 S protein in a closed conformation. Two fatty acids are found in each protomer of the HKU5 S protein, which stabilize the S protein in the closed conformation. Furthermore, we solve the structure of the HKU5 receptor-binding domain (RBD) in complex with the peptidase domain (PD) of Pipistrellus abramus angiotensin-converting enzyme 2 (ACE2), uncovering a unique binding mode distinct from other coronaviruses that use ACE2 as their receptor. Evolutionary and functional analyses indicate that mutations in the RBD can modulate receptor binding, while conservation and structural modeling suggest that HKU5 has the potential to cross the species barrier. Notably, we identify ACE2 orthologs in avian species, such as Pitta sordida, that support stable HKU5 RBD binding and interaction. Our functional assays, including pseudovirus entry and cell-cell fusion experiments, demonstrate that HKU5 can exploit ACE2 orthologs across species, providing molecular insights into its host adaptation and underscoring the importance of surveillance for this virus and its zoonotic risk.

biophysics↗

bioRxiv2025

MISO Regulates Mitochondria Dynamics and mtDNA Homeostasis by Establishing Inner Membrane Subdomains

Mitochondrial dynamics and mtDNA homeostasis are essential for numerous biological processes and have recently been linked through distinct subdomains termed small MTFP1-enriched mitochondria (SMEM). However, the molecular mechanisms governing this connection remain unclear. Here, we identified MISO (Mitochondrial Inner membrane Subdomain Organizer), a novel protein that regulates mitochondrial dynamics and drives SMEM formation in response to inner mitochondrial membrane (IMM) stresses. We demonstrate that MISO regulates mitochondrial dynamics in both Drosophila stem cells in vivo and mammalian cells in vitro. Mechanistic study showed that MISO promotes mitochondrial fission while inhibiting fusion. Notably, MISO induces SMEM formation and facilitates their peripheral fission, which regulates the lysosomal degradation of mtDNA. Meanwhile, MISO knock-out completely abolishes SMEM generation, indicating that MISO is both necessary and sufficient for SMEM biogenesis. Furthermore, SMEM functionality requires MISO-dependent recruitment of MTFP1 and subsequent engagement of downstream effectors, including the FIS1-DRP1 fission machinery. IMM stresses, including damages in mtDNA, OXPHOS complexes, and cristae, stabilize the normally short-lived MISO protein, thereby triggering SMEM assembly. Furthermore, MISO-orchestrated SMEM formation depends on its C-terminal domain, likely mediated by oligomerization. Together, our work elucidates a molecular mechanism by which IMM stresses affect mitochondrial dynamics and mtDNA homeostasis.

cell biology↗

bioRxiv2025

The microtubule-severing enzyme spastin regulates spindle dynamics to promote chromosome segregation in Trypanosoma brucei

Microtubule-severing enzymes play essential roles in regulating diverse cellular processes, including mitosis and cytokinesis, by modulating microtubule dynamics. In the early branching protozoan parasite Trypanosoma brucei, microtubule-severing enzymes are involved in cytokinesis and flagellum length control during different life cycle stages, but none of them have been found to regulate mitosis in any life cycle form. Here, we report the biochemical and functional characterization of the microtubule-severing enzyme spastin in the procyclic form of T. brucei. We demonstrate that spastin catalyzes microtubule severing in vitro and ectopic overexpression of spastin disrupts spindle microtubules in vivo in trypanosome cells, leading to defective chromosome segregation. Knockdown of spastin impairs spindle integrity and disrupts chromosome alignment in metaphase and chromosome segregation in anaphase. We further show that the function of spastin requires the catalytic AAA-ATPase domain, the microtubule-binding domain, and the microtubule interacting and trafficking domain, and that the association of spastin with spindle depends on the microtubule-binding domain. Together, these results uncover an essential role for spastin in chromosome segregation by regulating spindle dynamics in this unicellular eukaryote.

cell biology↗

bioRxiv2024

Exploring the specific predictive ability of multiple domains of spatial ability on STEM educational outcomes

AbstractResearch has implicated spatial ability as a robust predictor of aptitude, interest, and choice in STEM education and career pursuits. We address three under-explored questions regarding the role of spatial ability in STEM. First, can spatial ability consistently predict STEM success beyond other cognitive skills? Second, what aspects of spatial ability, if any, can predict success in STEM more accurately? Third, to what extent can genetic and environmental factors account for these predictions? We addressed these questions by leveraging data from the Twins Early Development Study (N = 3,936; age range = 16-22) and using 16 tests that assessed three domains of spatial ability: navigation, object manipulation, and visualization. Results show that all three domains are highly predictive of STEM educational outcomes, especially STEM degree choice. These associations persisted after accounting for verbal and general cognitive abilities (g), albeit attenuated. Associations were strongest for tests of object manipulation (e.g., 2D and 3D drawing, pattern assembly and mental rotation). Genetic factors accounted for most of the observed associations between spatial ability and STEM outcomes (62% - 86%) --genetic variance was mostly shared with g ([~] 40%) and, to a lesser extent, verbal ability ([~] 25%). Our findings highlight the potential utility of spatial ability as a specific predictor of success in STEM education and career choice beyond other cognitive abilities. Screening for and training spatial skills is likely useful for identifying potential, fostering talent, and improving outcomes in STEM. Significance StatementLeveraging a comprehensive battery of 16 spatial ability tests across multiple domains, we show that spatial ability has specific utility for predicting success in STEM education. Spatial skills predict success in STEM above and beyond other cognitive abilities, particularly when it comes to STEM engagement and pursuing further STEM education. Thus, spatial skills may be a fruitful target for policymakers, stakeholders, and industries looking to develop interventions, identify and foster talent, and reduce outcome disparities.

genetics↗

bioRxiv2024

Turncoat antibodies unmasked in a model of autoimmune demyelination: from biology to therapy

Autoantibodies contribute to many autoimmune diseases, yet there is no approved therapy to neutralize them selectively. A popular mouse model, experimental autoimmune encephalomyelitis (EAE), could serve to develop such a therapy, provided we can better understand the nature and importance of the autoantibodies involved. Here we report the discovery of autoantibody-secreting extrafollicular plasmablasts in EAE induced with specific myelin oligodendrocyte glycoprotein (MOG) antigens. Single-cell RNA sequencing reveals that these cells produce non-affinity-matured IgG antibodies. These include pathogenic antibodies competing for shared binding space on MOGs extracellular domain. Interestingly, the synthetic anti-MOG antibody 8-18C5 can prevent the binding of pathogenic antibodies from either EAE mice or people with MOG antibody disease (MOGAD). Moreover, an 8-18C5 variant carrying the NNAS mutation, which inactivates its effector functions, can reduce EAE severity and promote functional recovery. In brief, this study provides not only a comprehensive characterization of the humoral response in EAE models, but also a proof of concept for a novel therapy to antagonize pathogenic anti-MOG antibodies.

immunology↗

bioRxiv2024

SARM1 Activation Promotes Axonal Degeneration Via a Two-Step Liquid-to-Solid Phase Transition

SARM1 protein plays a central role in axonal degeneration, a key process in many neurodegenerative diseases and nerve injuries. It mediates this by depleting axonal NAD+ through its NADase activity, catalyzed by the TIR domain. Normally, this activity is kept in an inactive state and becomes activated in response to various neuronal damage signals. However, the molecular mechanism behind SARM1 activation, particularly how activation is restricted to damaged axons, remains to be fully elucidated. In this study, using a class of pyridine-containing compounds that induce SARM1-dependent cell death and axonal degeneration, we reveal a two-step process of SARM1 activation. The first step involves sub-lethal "priming" activation of SARM1 mediated by the NAD+ precursor NMN, leading to the formation of covalent conjugates between the hydrolyzed product of NAD+, adenosine diphosphate ribose (ADPR), and the compounds catalyzed by the intrinsic base exchange activity of SARM1. In the second step, these ADPR-conjugates act as molecular glues, promoting the formation of self-proliferating superhelical filaments. On these filaments, SARM1s TIR domains adopt an active NADase configuration. As the superhelical filaments rapidly reach their limit of solubility, they precipitate out of the liquid phase as condensates with stable, fully activated NAD hydrolysis activity. Interestingly, we found that a series of reported SARM1 inhibitors currently under clinical development, which target the TIR enzymatic domain, can paradoxically activate SARM1s NADase activity via this mechanism. These findings provide new insights into how SARM1 activation is spatially restricted to damaged axons and offer important implications for the development of therapeutics targeting SARM1.

molecular biology↗

bioRxiv2024

Alveolar macrophages initiate the spatially targeted neutrophil recruitment during nanoparticle inhalation

Exposure to air pollution, including nanoparticles (NPs), is a major health concern associated with various diseases, triggered by subtle inflammatory responses in the lung. To investigate the dynamic immune response in vivo, lung intravital microscopy (L-IVM), was used to analyze the behavior of alveolar macrophages (AMs) and neutrophils, combined with ventilator-assisted inhalation of nebulized NPs in mice. Inhalation of fluorescent quantum dot NPs (cQDs) and soot-like carbon black NPs (CNPs, ambient pollutants), led to rapid spatially focused recruitment of neutrophils near alveolar deposited NPs. Neutrophil recruitment was initiated by NPs uptake by AMs, dependent on AM motility and AM NP surface recognition. Prior airway application of neutralizing antibodies against alveolar ICAM-1 and LFA-1, leading to reduced AM motility, inhibition of C5aR1 and Fc{gamma}RI receptor mediated NPs uptake by AMs, as well as neutralizing of TNF and application of a cellular degranulation inhibitor, abolished the early immune response induced by NPs. Overall, our data demonstrates the crucial role of AM activity (migration, phagocytosis, cytokine release) in the rapid and site-specific recruitment of neutrophils during the early phase of particle inhalation, suggesting these processes to be key events in mounting the immune response upon NP inhalation in the lung. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=175 HEIGHT=200 SRC="FIGDIR/small/623349v1_ufig1.gif" ALT="Figure 1"> View larger version (52K): org.highwire.dtl.DTLVardef@d67e38org.highwire.dtl.DTLVardef@1f8c139org.highwire.dtl.DTLVardef@55be9dorg.highwire.dtl.DTLVardef@1555042_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2024

Circular RNA AEBP2: A novel regulator in dendritic cell development

Dendritic cells (DCs) are the main antigen-presenting cells and play a central role in activating adaptive immunity. How DCs is regulated during their development remains to be further elucidated. Circular RNA (circRNA) is a new type of non-coding RNA playing a critical in various cell events. However, little is known about the role of circRNA in DCs. Using in situ RNA hybridization, we found that DCs expressed exonic circular RNA AEBP2 (circAEBP2). circAEBP2 was upregulated in mature DCs compared to immature DCs. Knockdown of circAEBP2 in DCs using small interference RNA (siRNA) arrested DCs at an immature state with lower expression of MHC class II, CD40, CD80, and CD83. circAEBP2 knockdown in DCs also decreased the phosphorylation of nuclear factor-kappa B p65 (P-p65) but increased CD200R1. CircAEBP2-silenced DCs had a lower capacity to activate allogeneic CD4+ and CD8+ T cells in vitro and induced differentiation of regulatory T cells (Tregs). RNA immunoprecipitation assays revealed that circAEBP2 interacts with heterogeneous nuclear ribonucleoprotein F (hnRNP F) which is highly expressed in DCs and subsequently prevented hnRNP F nuclear translocation. In conclusion, we demonstrate that circAEBP2 regulates the development and function of DCs and knockdown of circAEBP2 induces tolerogenic DCs (Tol-DCs) through interacting with the hnRNP F/CD200R1/P-p65 signaling pathway. circAEBP2 is a new regulator in DCs and modulating it provides a novel strategy for inducing Tol-DCs.

immunology↗

bioRxiv2024

Key contribution of prefrontal inhibition to passive coping behaviour: chronic stress and fast-acting antidepressant

Persistent passive coping behaviour is a hallmark feature in major depression and is reversed by fast-acting antidepressants (such as ketamine). This behaviour is regulated by a specific cortico-midbrain circuit. However, whether the prefrontal cortex (PFC), especially inhibition in PFC, contributes to the modulation of passive coping, and whether this modulation is important for mediating the impacts of chronic stress and/or fast-acting antidepressants, are poorly understood. Here, we found that rostral prelimbic cortex (rPL) bidirectionally controls the occurrence of passive coping behaviour where excitatory and inhibitory neurons play opposite roles. Chronic stress leads to reduced excitation/inhibition (E/I) ratio, reflected as alterations in in vivo spiking rate, synaptic inputs and intrinsic excitability of both excitatory and inhibitory neurons. A fast-acting antidepressant, (2R, 6R)-hydroxynorketamine (HNK), reduced passive coping behaviour, restored rPL E/I ratio and partially reversed altered properties in rPL neurons, in chronically stressed mice. Importantly, chronic stress and HNK mostly affected fast-spiking/parvalbumin inhibitory neurons instead of other inhibitory neurons, indicating the important role of this subtype of inhibitory neurons in the above processes. These findings demonstrate the importance of rPL E/I balance in regulating passive coping, which can be modulated by chronic stress and rapidly restored by fast-acting antidepressant.

neuroscience↗

bioRxiv2024

Nitroxoline evidence Amoebicidal Activity against Acanthamoeba castellanii through DNA damage and the stress response pathways

Acanthamoeba castellanii is a widespread unicellular eukaryote found in diverse environments, including tap water, soil, and swimming pools. It is responsible for severe infections, such as Acanthamoeba keratitis and granulomatous amebic encephalitis, particularly in individuals with immunocompromisation. The ability of protozoans to form dormant cysts complicates treatment, as current therapies are ineffective against cyst stages and suffer from poor specificity and side effects. Nitroxoline, a quinoline derivative with well-established antibacterial, antifungal, and antiviral properties, is a promising therapeutic candidate. This study aimed to elucidate cellular signalling events that counteract the effects of nitroxoline. In this study, nitroxoline significantly reduced the viability of A. castellanii trophozoites in a dose- and time-dependent manner, inducing morphological changes and apoptosis. Transcriptomic analysis revealed substantial alterations in gene expression, including enrichment of metabolic pathways, DNA damage responses, and iron ion binding. Nitroxoline treatment upregulated genes involved in DNA repair and oxidative stress response while regulating genes in the methionine and cysteine cycles. It also decreased the mitochondrial membrane potential, H2S production, and total iron amount in A. castellanii. Bioinformatics analyses and molecular docking studies suggest direct interactions between nitroxoline and A. castellanii proteins. Our research provides a comprehensive molecular map of the response of A. castellanii to nitroxoline, revealing significant changes in gene expression related to the stress response and metabolic pathways. These findings underscore the potential of nitroxoline as a potent anti-Acanthamoeba agent, offering new insights into its mechanism of action and paving the way for effective therapeutic strategies.

microbiology↗

bioRxiv2024

Biologically-informed Killer cell immunoglobulin-like receptor (KIR) gene annotation tool

Natural killer (NK) cells are essential components of the innate immune system, with their activity significantly regulated by Killer cell Immunoglobulin-like Receptors (KIRs). The diversity and structural complexity of KIR genes present significant challenges for accurate genotyping, essential for understanding NK cell functions and their implications in health and disease. Traditional genotyping methods struggle with the variable nature of KIR genes, leading to inaccuracies that can impede immunogenetic research. These challenges extend to high-quality phased assemblies, which have been recently popularized by the Human Pangenome Consortium. This paper introduces BAKIR (Biologically-informed Annotator for KIR locus), a tailored computational tool designed to overcome the challenges of KIR genotyping and annotation on high-quality, phased genome assemblies. BAKIR aims to enhance the accuracy of KIR gene annotations by structuring its annotation pipeline around identifying key functional mutations, thereby improving the identification and subsequent relevance of gene and allele calls. It uses a multi-stage mapping, alignment, and variant calling process to ensure high-precision gene and allele identification, while also maintaining high recall for sequences that are significantly mutated or truncated relative to the known allele database. BAKIR has been evaluated on a subset of the HPRC assemblies, where BAKIR was able to improve many of the associated annotations and call novel variants. BAKIR is freely available on GitHub, offering ease of access and use through multiple installation methods, including pip, conda, and singularity container, and is equipped with a user-friendly command-line interface, thereby promoting its adoption in the scientific community.

genomics↗

bioRxiv2024

AmygdalaGo-BOLT3D: A boundary learning transformer for tracing human amygdala

Each year, thousands of brain MRI scans are collected to study structural development in children and adolescents. However, the amygdala, a particularly small and complex structure, remains difficult to segment reliably, especially in developing populations where its volume is even smaller. To address this challenge, we developed AmygdalaGo-BOLT, a boundary-aware deep learning model tailored for human amygdala segmentation. It was trained and validated using 854 manually labeled scans from pediatric datasets, with independent samples used to ensure performance generalizability. The model integrates multiscale image features, spatial priors, and self-attention mechanisms within a compact encoder-decoder architecture to enhance boundary detection. Validation across multiple imaging centers and age groups shows that AmygdalaGo-BOLT closely matches expert manual labels, improves processing efficiency, and outperforms existing tools in accuracy. This enables robust and scalable analysis of amygdala morphology in developmental neuroimaging studies where manual tracing is impractical. To support open and reproducible science, we publicly release both the labeled datasets and the full source code. TeaserAn AI model enables accurate amygdala segmentation across ages and centers, supporting large-scale brain imaging studies.

neuroscience↗