Biology subjects

Chang, J.

Publications and source records attributed to Chang, J..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

Structural semantic evolutionary distance (SSED) unifies the selection of cancer driver genes across macroevolution and tumorigenesis.

The non-random, site-specific enrichment of somatic mutations in cancer driver genes (CDGs) suggests their emergence is governed by underlying evolutionary constraints. However, quantitative methods to define these constraints and their underlying principles remain underexplored. To address this, we introduce the Structural Semantic Evolutionary Distance (SSED), a metric leveraging the pretrained ESM-3 protein language model to quantify evolutionary divergence within a unified structural semantic space. Our analysis demonstrates that CDGs are subject to persistent structural semantic constraints across species, tolerating a significantly narrower range of structural semantic changes during evolution compared to non-CDGs. Crucially, clinically observed oncogenic mutations follow this same principle, favoring minimal structural perturbation as shaped by long-term gene evolution. Such mutations maintain core protein function while conferring a capacity for immune evasion, thereby driving clonal expansion. Guided by this "evolutionary constraint" framework, we successfully predicted and experimentally validated a previously uncharacterized oncogenic mutation, KRAS R135L, in bronchial epithelial cells. Furthermore, clinical cohort analysis demonstrated that SSED acts as an independent predictor of response to immune checkpoint blockade, offering information orthogonal to tumor mutational burden (TMB). This study unifies the evolutionary principles governing CDGs across macroevolutionary and microevolutionary (tumorigenesis) timescales, elucidates the balance between structural adaptability, functional conservation, and immune pressure, and identifies a novel predictive biomarker for cancer immunotherapy.

genetics↗

bioRxiv2025

A K27-linked Ubiquitin Checkpoint Controls NOTCH Homeostasis

Cell surface receptors such as NOTCH1 must be tightly regulated to ensure developmental fidelity and prevent pathological activation. Although the proteolytic steps culminating in nuclear NOTCH1 signaling are established, how cells prevent excessive or uncontrolled activation has remained unresolved. Here we identify the autophagy-related protein UVRAG as a negative regulator of NOTCH1. Upon receptor activation, UVRAG, acting independently of autophagy, recruits and activates the E3-ligase ITCH to catalyze K27-linked ubiquitination of membrane-tethered NOTCH1, thereby licensing ESCRT-dependent lysosomal degradation. Disruption of the UVRAG-ITCH-ESCRT axis stabilizes activated NOTCH1 intermediates and amplifies oncogenic signaling. In T-cell leukemia models driven by constitutive NOTCH1 activity, restoring UVRAG expression reinstates receptor turnover, suppresses disease progression, and improves therapeutic response. These findings define a ubiquitin-directed safeguard circuit that enforces NOTCH1 signaling homeostasis and reveals a tunable axis for intervention in NOTCH1-driven cancers.

molecular biology↗

bioRxiv2025

Signaling induced biophysical disruption of repressed chromatin domains drives immune cell fate

Cell fate transitions require signal-induced chromatin derepression, yet mechanisms governing transitions from repressed to active chromatin states are poorly understood. We discover, at fate-defining genes across immune cell types, a signal-induced histone code, and describe domains of H3 serine 28 phosphorylation (H3S28ph) spanning architectural features, often coincident with repressive H3 lysine 27 trimethylation (H3K27me3). Employing biophysical, single cell, and functional approaches to study signal-induced cell differentiation in the immune system, we uncover epigenomic transitions and cell fate choices precipitated by histone phosphorylation (H3ph). Mechanistically, H3ph overrides Polycomb Repressive Complex 2 (PRC2) chromatin repression, biophysically disrupts polynucleosome compaction, and promotes loss of H3K27me3, while increasing activating H3K27 acetylation and H3K36 dimethylation to drive domain interactivity and stabilize transcription. We demonstrate the activity of H3ph in several cell fate transitions and illuminate biophysical mechanisms enabling rapid signal-activated chromatin derepression, processes with general relevance for cellular differentiation and activation.

immunology↗

bioRxiv2025

Histology-informed spatial domain identification through multi-view graph convolutional networks

Identifying spatial domains is crucial in spatial transcriptomics, yet effectively integrating gene expression, spatial location, and histology remains challenging. We present STESH, a Spatial Transcriptomics clustering method that combines Expression, Spatial information and Histology. STESH extracts histological features using a convolutional neural network and generates expression, histology, spatial, and collaborative convolution modules for a multi-view graph convolutional network with a decoder and attention mechanism. We evaluated STESH on multiple tissue types and technology platforms. STESH consistently outperformed ten state-of-the-art methods, achieving superior clustering accuracy with the highest scores in adjusted Rand index, normalized mutual information, and Fowlkes-Mallows index.

bioinformatics↗

bioRxiv2025

Silencing the signal: The metastasis suppressor NDRG1 disrupts exosome-mediated crosstalk in pancreatic cancer

Pancreatic cancer (PaC) remains one of the deadliest cancers, with 5-year survival rates of 13%. A major driver of its aggressiveness is the tumour microenvironment (TME), which fuels tumour growth, metastasis, and therapeutic resistance through dynamic, bi-directional communication between cancer cells, fibroblasts, and immune cells. Emerging evidence highlights extracellular vesicles (EVs) as key mediators of oncogenic cross-talk within the PaC TME. This study demonstrates for the first time that the metastasis suppressor NDRG1 significantly influences the biogenesis, cargo packaging and release of EVs by cancer cells. This was mediated by a direct interaction between NDRG1 and ALIX, a key protein involved in EV biogenesis and packaging, with NDRG1 facilitating ALIX proteasomal degradation. Further, EVs released from NDRG1-overexpressing cells had significantly fewer CAF-activation proteins (i.e. TGF-B), leading to attenuated ERKl/2 and p38 activation in pancreatic stellate cells (PSCs), and reduced expression of key fibrotic markers (a-SMA, FAP, and collagen 1A). NDRG1 also reduced EV uptake by PaC cells and diverted these to the lysosome for degradation. These findings uncover a previously unrecognized mechanism by which NDRG1 disrupts the oncogenic two-way communication between PaC cells and the TME, positioning NDRG1 as a compelling therapeutic target against this formidable malignancy. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=123 SRC="FIGDIR/small/682724v1_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@154b908org.highwire.dtl.DTLVardef@13fe6f5org.highwire.dtl.DTLVardef@540fadorg.highwire.dtl.DTLVardef@d15636_HPS_FORMAT_FIGEXP M_FIG C_FIG

cancer biology↗

bioRxiv2025

Differential colocalization of cell-type pairs in single-cell spatial omics with DC-SPOMIC

MotivationSpatial omics studies compare cell-cell organization across samples, but most methods model between-sample variability while treating sample-level spatial estimates as error-free. Overlooking within-sample uncertainty can distort inference in heterogeneous cohorts, motivating methods that explicitly quantify and propagate this uncertainty into cohort-level analyses. ResultsWe present PANORAMIC, a hierarchical framework for spatial colocalization analysis that uses edge-corrected neighborhood enrichment to estimate local cell-type colocalization, spatial bootstrapping to quantify within-sample uncertainty, and multilevel random-effects meta-analysis to propagate this uncertainty across samples, patients, and conditions. In simulations, PANORAMIC improved recovery of within-sample uncertainty and between-sample heterogeneity relative to naive estimators across diverse spatial settings and progressive data degradation. Applied to a colorectal cancer tissue microarray profiled by multiplexed immunofluorescence imaging, PANORAMIC identified stronger B- and T-cell colocalization in tumors with Crohns-like reaction than in tumors with diffuse inflammatory infiltration, together with tighter higher-order immune organization consistent with immune aggregates. These results show that propagating within-sample spatial uncertainty can improve cohort-level inference in spatial omics studies. Availability and ImplementationPANORAMIC is released as an open-source R package at https://github.com/plevritis-lab/panoramic. Contact[sylvia.plevritis@stanford.edu]

bioinformatics↗

bioRxiv2025

Bioactive 3D-bioprinted scaffolds with endothelial progenitor cells and Zn2+-enhanced SFMA/HisMA/nHAP for healing of infected bone defect

Infected bone defects remain one of the greatest challenges in orthopedics, as bacterial contamination and insufficient vascularization severely compromise regeneration. Here, we report the development of a multifunctional 3D-bioprinted scaffold composed of zinc ion (Zn{superscript 2}{square}) functionalized silk fibroin methacryloyl (SFMA), histidine methacryloyl (HisMA), and nano-hydroxyapatite (nHAP), further loaded with endothelial progenitor cells (EPCs) to promote simultaneous antibacterial, angiogenic, and osteogenic responses. The photocrosslinkable SFMA/HisMA bio-ink reinforced with nHAP provided mechanical stability, controlled swelling and degradation, while Zn{superscript 2}{square} coordination endowed strong antibacterial activity against E. coli and S. aureus. EPCs adhered and proliferated on the scaffolds, and in co-culture with bone marrow stromal cells (BMSCs) markedly enhanced osteogenic differentiation, as evidenced by increased expression of OCN, RUNX2, COL1, and OPN. Proteomic profiling further revealed that EPC-loaded scaffolds modulated osteoblast protein expression patterns consistent with both immunoregulation and active extracellular matrix reconstruction to promote tissue regeneration. In a murine femoral infected defect model, EPC-loaded scaffolds suppressed pro-inflammatory cytokines (IL-6, TNF-), stimulated angiogenesis, and supported robust new bone formation, leading to accelerated defect repair confirmed by {micro}CT and histological analyses. Together, these findings demonstrate that the EPC-loaded Zn{superscript 2}{square} functionalized SFMA/HisMA/nHAP scaffold integrates antibacterial defence with vascular and osteogenic stimulation, offering a promising translational strategy for the treatment of complex infected bone defects. Graphic Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=145 SRC="FIGDIR/small/676384v1_ufig1.gif" ALT="Figure 1"> View larger version (51K): org.highwire.dtl.DTLVardef@1a75a57org.highwire.dtl.DTLVardef@1c32b6org.highwire.dtl.DTLVardef@1456159org.highwire.dtl.DTLVardef@1fa5321_HPS_FORMAT_FIGEXP M_FIG C_FIG Schematic illustration of materials synthesis and application. Silk fibroin was modified with glycidyl methacrylate (GMA) to obtain silk fibroin methacryloyl (SFMA), while histidine was functionalized with acryloyl chloride to generate histidine methacryloyl (HisMA). SFMA and HisMA were then combined and coordinated with Zn{superscript 2}{square} ions and incorporated with nHAP to form the full-component bioactive scaffold (SFMA/HisMA/nHAP/Zn{superscript 2}{square}). The resulting material can be 3D-printed and loaded with endothelial progenitor cells (EPCs) to yield a multifunctional scaffold with antibacterial activity, pro-osteogenic and pro-angiogenic properties, favorable physicochemical characteristics, and excellent biocompatibility, making it a promising candidate for the treatment of infected bone defects.

bioengineering↗

bioRxiv2025

A Rapid and Scalable Subcutaneously Administered Murine Thymus Micro-organoid for Generating Functional T cells

Thymic function can decline due to age-related involution, congenital disorders, acute infections, or chemo/radiation therapy. Decline in thymic function leads to decreased T cell production and weakened immunity. To address these thymic insufficiencies, we aimed to develop a transplantable and scalable micro-organoid system utilizing fibroblasts and thymic cells. We have developed a reliable and rapid method to generate thymic micro-organoids using selectively screened fibroblasts and murine thymic cells. The thymic micro-organoids are cryo-preservable, injectable, and give rise to T cells both in vitro and in vivo. Thymic organoids expressed key genes required to sustain T cell development and maturation: ccl25, dll-1, dll-4, foxn-1, il-7, scf. When injected into T cell-deficient Prkdcscid mice, the organoids gave rise to functional {beta}, {psi}8, natural killer T (NKT) cells, and FoxP3+ regulatory T cells. Organoid-derived T cells expressed a diverse T cell receptor (TCR) repertoire in vivo and responded to stimulation with anti-CD3/28, Concanavalin-A, or Phytohemagglutinin. Thymic organoids derived from pmel-1 thymocytes gave rise to V{beta}13+ T cells that delayed the growth of B16 melanoma and enhanced activation of T and NK cells. This approach presents a valuable tool for mechanistic studies and addressing current therapeutic gaps in diseases associated with thymic decline and insufficiencies.

immunology↗

bioRxiv2025

Integrated histopathologic modeling of detailed tumor subtypes and actionable biomarkers

Accurate cancer subtyping with accompanying molecular characterization is critical for precision oncology. While machine learning approaches have been applied to both digital pathology and cancer genomics, previous work has been limited in sample size and has typically aggregated granular cancer subtypes into coarse groupings, likely obfuscating informative molecular and prognostic associations and phenotypic variation of more detailed tumor subtypes. Accordingly, we collated 378,123 hematoxylin and eosin (H&E)-stained whole-slide images (WSIs) with matched targeted DNA clinical sequencing results and OncoTree detailed cancer subtypes from a real-world cohort of 71,142 patients. Using this scaled, granular dataset and a cancer subtype knowledge graph, we developed Mosaic: a family of calibrated machine learning models using H&E WSI embeddings to classify tumors and identify molecular phenotypes across 163 detailed subtypes. The cancer subtyping module (Aeon) achieved an area under the receiver operating characteristic curve (AUROC) of 0.992 overall, with 161/163 subtypes reaching an AUROC [≥] 0.90 and improved performance over a state-of-the-art genomics-based classifier. The genomic inference module (Paladin) achieved an AUROC [≥] 0.80 for 167 pairs of detailed subtypes and genomic targets. We further used the learned histopathologic representations to i) identify key associations of the histopathologic embeddings with clinical biomarkers; ii) identify unsupervised sub-clusters of tumors with genomic determinants of tumor phenotype; iii) specify granular diagnoses for cancers of unknown primary, evaluated by genomic associations and expected clinical outcome distributions; iv) annotate functional significance for variants of uncertain significance (VUS); and v) identify cases that mimic the phenotypic effect of known DNA variants on H&E in the absence of detectable DNA alterations. Taken together, this work advances our understanding of phenotypic variation of granular tumor subtypes, their relevance to enhanced diagnostics, and their potential utility in risk stratification with multimodal machine learning in cancer.

cancer biology↗

bioRxiv2025

Distinct myeloid precursors and their interaction with mesenchymal cells orchestrate the spreading of psoriatic disease from the skin to the joints

Psoriatic disease initially affects the skin, but later extends to the joints. Herein, we describe a two-step process that orchestrates spreading of inflammation from the skin to the joints. Induction of psoriatic skin disease in photoconvertible mice, followed by sequencing and computational characterization of skin-derived cells in the joints, identified a unique population of CD2+ MHC-II+ CCR2+ myeloid precursors that built up a skin-derived myeloid cell compartment in the joints. Single-cell cross-species reference mapping and mitochondrial variant tracing showed an orthologue human cell population. Interactome analyses in the joints showed that in a second step, resident regulatory CD200+ fibroblasts critically regulate the priming of the CD2+ MHC-II+ CCR2+ myeloid precursors, which subsequently regulate the IL-17 expression in T cells. Hence, spreading of inflammation requires a distinct migratory myeloid precursor population and a permissive local tissue environment, similar to tumour metastasis. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=121 SRC="FIGDIR/small/665302v1_ufig1.gif" ALT="Figure 1"> View larger version (24K): org.highwire.dtl.DTLVardef@38c440org.highwire.dtl.DTLVardef@dc574dorg.highwire.dtl.DTLVardef@1e79f72org.highwire.dtl.DTLVardef@19508cf_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2025

Autism Mutations Preferentially Target Distributed Brain Circuits and Cell Types Associated with Sensory Integration and Decision Making

Autism spectrum disorder (ASD) is a highly heritable psychiatric condition characterized by difficulties in social communication and stereotypic, repetitive behaviors. Genetics studies have discovered many dozens of genes with causal roles in autism, and functional analyses have demonstrated that ASD-associated mutations affect a diverse range of brain regions and cell types. However, the precise mechanisms by which these genetic alterations lead to autism-related phenotypes remain unclear. Psychiatric cognitive and behavioral traits are believed to arise from dysfunction of specific brain circuits formed by anatomically and functionally connected brain areas. To identify the circuits and cell types primarily affected by ASD mutations, we developed an unbiased approach, GENCIC, which computationally integrates genome-wide genetic data with a brain-wide spatial mouse transcriptome and an anatomical mesoscale connectome. Applying this approach to ASD reveals a convergence of mutations on cohesive brain circuits that are enriched in long-distance connections and span both cortical and subcortical brain structures. Furthermore, our analysis of brain-wide single-cell spatial transcriptomics shows that the heterogeneity of circuit structures affected in ASD is matched by the substantial diversity of strongly impacted circuit cell types. Notably, the implicated circuits and cell types play a central role in the integration of multimodal sensory and emotional information and in decision-making based on this information. We also find that different circuit structures exhibit distinct vulnerability patterns that correlate with cognitive phenotypes in ASD. Overall, our study demonstrates how ASD-related genetic mutations impact multiple levels of brain organization, ultimately disrupting functional circuits that drive core autism-related behaviors.

systems biology↗

bioRxiv2025

De Novo design of a potent Wnt Surrogate specific for the frizzled7 subtype members

In humans, 19 Wnt ligands interact with 10 Frizzled (Fzd) receptors and the co-receptors LRP5/6 to initiate signaling. Wnts and Fzds are highly promiscuous, making it challenging to dissect the specific outcomes of individual Wnt-Fzd interactions. Developing Wnt surrogates with specificity for individual Fzd subtypes could be pivotal. We present a modular, potent, and Fzd7-specific Wnt surrogate that consists of three de novo designed modules, a Fzd7 binder, an LRP6 binder and a homodimeric protein. The Fzd7-specific module was designed by targeting two less conserved surface patches on the cysteine-rich domain (CRD) of Fzds to achieve both selectivity and affinity. It exhibits a strong binding affinity (KD < 2.3 nM) for the very closely related Fzd7 subtype members (Fzd7, Fzd1, Fzd2) with no measurable binding to the CRDs of the other seven Fzd receptors. This Wnt surrogate induced spheroid organoid formation from intestinal stem cells at subnanomolar concentration, and promoted full hair follicle regeneration and robust hair growth in mice. These results suggest that our strategy could be extended to design modular Wnt surrogates capable of selectively activating individual Fzd receptors, providing a valuable tool kit for development and differentiation, organoid cultures and targeted regeneration.

bioengineering↗

bioRxiv2025

The Renin-Angiotensin System Modulates SARS-CoV-2 Entry via ACE2 Receptor

The RAS plays a central role in cardiovascular regulation and has gained prominence in the pathogenesis of COVID-19 due to the critical function of ACE2 as the entry receptor for SARS-CoV-2. Angiotensin IV, but not angiotensin II, has recently been reported to enhance the binding between the viral spike protein and ACE2. To investigate the virological significance of this effect, we developed a single-round infection assay using SARS-CoV-2 viral-like particles expressing the spike protein. Our results demonstrate that while angiotensin II does not affect viral infectivity across concentrations ranging from 40LnM to 400LnM, angiotensin IV enhances viral entry at a low concentration but exhibits dose-dependent inhibition at higher concentrations. These findings highlight the unique dual role of angiotensin IV in modulating SARS-CoV-2 entry. In silico molecular docking simulations indicate that angiotensin IV was predicted to associate with the S1 domain near the receptor-binding domain in the open spike conformation. Given that reported plasma concentrations of angiotensin IV range widely from 17 pM to 81 nM, these levels may be sufficient to promote, rather than inhibit, SARS-CoV-2 infection. This study identifies a novel link between RAS-derived peptides and SARS-CoV-2 infectivity, offering new insights into COVID-19 pathophysiology and informing potential therapeutic strategies.

microbiology↗

bioRxiv2025

Microengineered transplantation of human solid tumors for in vitro studies of CAR T immunotherapy

Treatment of solid malignancies using chimeric antigen receptor (CAR) T cells remains a significant challenge, but current efforts to advance this therapy are challenged by our limited capacity to probe and understand cancer-immune interactions in human solid tumors. Here, we present a microengineered platform for in vitro modeling of malignant solid tumors during CAR T therapy. This system makes it possible to vascularize human tumor explants and perfuse them with blood-borne immune cells in a controlled manner. We first present a microphysiological model of human lung adenocarcinomas infused with CAR-T cells and show how this system can be used to simulate, visualize, and interrogate tumor-directed trafficking and effector function of CAR T cells. We then demonstrate the proof-of-principle of testing a chemokine-directed CAR T cell engineering strategy in a model of malignant pleural mesothelioma and validating our in vitro assessment using a matching in vivo mouse model. Finally, we describe a potential therapeutic target discovered by single-cell RNA sequencing that can be pharmacologically modulated to increase the efficacy of CAR T cells for lung adenocarcinoma, for which we also present specific biomarkers identified by global metabolomics analysis. We believe that the bioengineering principle demonstrated here will make important contributions to developing new capabilities for preclinical studies of adoptive cell therapies for cancer and other complex diseases.

bioengineering↗

bioRxiv2025

Identifying potential biomarkers of response to CCR5Δ32 HSCT in HIV infection

While HIV can be effectively suppressed to a chronic, mostly asymptomatic infection with combination antiretroviral therapy (cART), a cure is still needed. Hematopoietic stem cell transplantation (HSCT) from HIV-resistant donors has shown promise and has resulted in HIV remission in five patients. However, this treatment strategy does not guarantee HIV remission; six other patients who received a similar transplant had poor outcomes and died within a year of treatment. These different outcomes may be due to inter-individual differences in HIV infection dynamics that result in heterogeneity of therapeutic responses to HSCT. Using a previously published mechanistic model of HIV infection and virtual populations calibrated from patient data, we performed simulations to understand how different parameters in the model can influence the observed heterogeneity in therapeutic outcomes across virtual patient populations. Our simulations confirmed that discontinuation of cART, without HSCT, always leads to viral rebound, and that time to rebound differs across patients due to the interindividual variability (IIV) in underlying infection dynamics. Extending the duration of cART only slightly increased the predicted median time to rebound and its variance. By contrast, HSCT followed by cART cessation led to HIV remission, but only for a subset of the virtual patients. The proportion of patients predicted to go into remission depends directly on the ratio of donor to host cell immune cells in the post-HSCT chimeric immune system. Of the mechanistic model parameters, no single parameter determined whether a patient was a responder or a non-responder; rather, the interactions between multiple model parameters were crucial in driving treatment responses. In contrast, virtual equivalents of clinically accessible observations, e.g. viral load and cell populations at specific times, were shown to be better predictors than mechanistic model parameters in separating patients into non-responding (viral rebound) and responding (no rebound) clusters. One Sentence SummarySimulations using a mechanistic HIV model suggest that interindividual variability (IIV) in infection dynamics drives differential responses to hematopoietic stem cell transplantation (HSCT), with HIV remission being largely dependent on donor-to-host immune cell ratios in the post-HSCT chimeric immune system and better predicted by clinically accessible observations than individual model parameters.

systems biology↗

bioRxiv2025

Neural signatures of motor memories emerge in neural network models

Animals can learn and seamlessly perform a great number of behaviors. However, it is unclear how neural activity can accommodate new behaviors without interfering with those an animal has already acquired. Recent studies in monkeys performing motor and brain-computer interface (BCI) learning tasks have identified neural signatures--so-called "memory traces" and "uniform shifts"--that appear in the neural activity of a familiar task after learning a new task. Here we asked when these signatures arise and how they are related to continual learning. By modeling a BCI learning paradigm, we show that both signatures emerge naturally as a consequence of learning, without requiring a specific mechanism. In general, memory traces and uniform shifts reflected savings by capturing how information from different tasks coexisted in the same neural activity patterns. Yet, although the properties of these two different signatures were both indicative of savings, they were uncorrelated with each other. When we added contextual inputs that separated the activity for the different tasks, these signatures decreased even when savings were maintained, demonstrating the challenges of defining a clear relationship between neural activity changes and continual learning.

neuroscience↗

bioRxiv2025

Rare mutations implicate CGE interneurons as a vulnerable axis of cognitive deficits across psychiatric disorders

Neuropsychiatric disorders such as autism spectrum disorder (ASD) and schizophrenia (SCZ) share genetic risk factors, including genes affected by rare high-penetrance single nucleotide variants (SNVs) and copy number variants (CNVs). ASD and SCZ exhibit both overlapping and distinct clinical phenotypes. Cognitive deficits and intellectual disability--critical predictors of long-term outcomes--are common to both conditions. To investigate shared and disorder-specific neurobiological impact of highly penetrant rare mutations in ASD and SCZ, we analyzed human single-nucleus whole-brain sequencing data to identify strongly affected brain cell types. Our analysis revealed caudal ganglionic eminence (CGE)-derived GABAergic interneurons as a key nexus for cognitive deficits across these disorders. Notably, genes within 22q11.2 deletions, known to confer a high risk for SCZ, ASD, and cognitive impairment, showed a strong expression bias toward vasoactive intestinal peptide-expressing cells (VIP+) among CGE subtypes. To explore perturbations of VIP+ GABAergic interneurons in the 22q11.2 deletion syndrome in vivo, we examined their activity in the Df(16)A+/- mouse model during a spatial navigation task and observed reduced activity along with altered responses to random rewards. At the population level, VIP+ interneurons exhibited impaired spatial encoding and diminished subtype-specific activity suggesting deficient disinhibition in CA1 microcircuits in the hippocampus, a region essential for learning and memory. Overall, these results demonstrate the crucial role of CGE-derived interneurons in mediating cognitive processes that are disrupted across a range of psychiatric and neurodevelopmental disorders.

neuroscience↗

bioRxiv2025

Assessing the combined performance of supervised learning and spike-in constructs for bias correction in eDNA metabarcoding

Environmental DNA (eDNA) metabarcoding has become increasingly popular as an approach to efficiently document biodiversity within an environment characterized by relative uncertainty. Compared to the traditional stereomicroscopic approaches, eDNA metabarcoding is simpler and less costly. Under ideal circumstances, researchers are able to directly extrapolate the true relative abundance of a particular taxon in the sampled environment by computing the proportion of sequenced reads assigned to the specific taxon. Although several previous studies have been carried out under such assumptions, some researchers have raised the possibility that there may exist both biological and technical biases in eDNA metabarcoding studies, leading to inconsistent estimations of community composition. Using mock community datasets from nine relevant studies in the past, we showed that bias correction in eDNA metabarcoding studies is indeed a predictable task. We also found reads and amp_gc to be the two most important feature predictors, such that these two features alone are enough to retain most of the model performances. Experiment-specific information were found to be necessary for bias correcting models to perform well. However, we have yet to develop an effective way of converting knowledge regarding spike-in (SP) samples into experiment-specific information that can be learned by existing models. Nonetheless, under the data-specific scenario, AdaBoost showed an optimal 35.62% improvement from the baseline established by the vanilla control model. Additionally, we showed that model performances could be rescued by the availability of experiment-specific data, under which XgBoost exhibited an optimal 81.57% improvement from the baseline. Our work suggests that future metabarcoding studies would benefit from performing supervised learning (SL)-based bias correction prior to downstream analyses. Moreover, if experiment-specific data is available at the time of the study, it is optimal to construct an XgBoost model. Otherwise, it is still recommended to construct an AdaBoost model, which showed marginal improvement from the baseline with no modeling. One Sentence SummarySupervised learning models, particularly XgBoost and AdaBoost, can effectively correct biases in eDNA metabarcoding studies, with performance improving significantly when experiment-specific data is available.

bioinformatics↗