Biology subjects

Gillis, J.

Publications and source records attributed to Gillis, J..

At least 19 recordsLinked to original sources

bioRxiv2026

Detecting cell segmentation errors using doublet methods

In spatial transcriptomics, cell segmentation is used to draw boundaries around cells. Molecules located within a cell's boundary are assigned to it, making its gene expression profile dependent on segmentation accuracy. To identify potentially problematic cells, studies increasingly use doublet detection methods, a class of algorithms developed to recognize molecular admixture from two cells in scRNA-seq. To evaluate their suitability in spatial data, we model cell segmentation errors as a continuum of partial molecular admixture between neighboring cells, generated by varying the loss of a cell's own transcripts and the gain of transcripts from its neighbor. Evaluating 8 doublet methods across 16 spatial datasets, we characterize the conditions under which they perform well and identify their failure modes. Detection improves with increasing molecular admixture and transcriptional dissimilarity between neighboring cells, with cxds2, scDblFinder.score, and a simple baseline (unique_genes) performing best. These patterns are consistent across datasets spanning different gene panels, technological platforms, staining techniques, segmentation algorithms, and error frequencies. In unperturbed spatial data, elevated doublet scores localize to regions consistent with segmentation problems, suggesting that they can help prioritize cells or regions for inspection, segmentation refinement, or transcript reassignment. Our results establish when doublet methods can provide useful quality control signals for cell segmentation, supporting their use in spatial transcriptomics.

bioinformatics↗

bioRxiv2026

Neuronal and vascular genome-maintenance states organize opposing multicellular architectures in the aging brain

Genome maintenance is usually treated as cell-intrinsic, yet brain cells age within multicellular neighborhoods. Whether a cell's DNA damage response (DDR) state is systematically embedded in native tissue architecture is unknown. We used MERFISH to map curated repair programs across 604,252 cells in matched sections from 2-3 and 22-month-old mice. Regional DDR expression and coordination formed an age-dependent mosaic, accompanied by immune and oligodendroglial shifts. Independent metacell and distance-resolved analyses identified a consistent anchor-dependent organization. Low-DDR neurons occupied immune- and OPC-oligodendroglial-rich niches. In contrast, DDR-high vascular anchors were associated with vascular-cell enrichment and, in cortex and cerebral nuclei, OPC-oligodendroglial enrichment, whereas immune cells were enriched around DDR-low vascular anchors. Thus, genome-maintenance state was associated with distinct anchor-cell-specific spatial context depending on anchor-cell identity and its DDR transcriptional state. These opposing neuronal and vascular patterns define anchor-dependent genome-maintenance niches as a spatial feature of brain aging, with implications for regional vulnerability.

neuroscience↗

bioRxiv2026

Robustness to nuisance perturbations enables unsupervised evaluation of single-cell foundation models

Single-cell foundation model evaluations have relied almost exclusively on downstream tasks. While these tasks measure whether an embedding recovers annotated cell types, batches, or trajectories, they cannot determine if that structure is reproducible or merely an artifact of a single noisy draw, a key limitation since incomplete sampling is intrinsic to single-cell measurement. Here, we introduce a fully unsupervised evaluation framework grounded in a fundamental principle: a faithful representation must preserve its neighbourhood structure under nuisance perturbations that mimic technical and sampling variation. Across five scFMs, a PCA baseline, and 39 datasets, we show that models ranked as near-equivalent by standard benchmarks differ nearly twofold in local neighbourhood preservation under a perturbation discarding just 5% of counts. This structural instability is scale-dependent and often masked by visually coherent embeddings. Cluster-level stability under resampling tracks established bio-conservation metrics (Spearman {rho}=0.78), showing that invariance to nuisance perturbations captures representation quality no benchmark measures directly.

bioinformatics↗

bioRxiv2026

Genetic and stochastic basis of phenotypic discordance in 16p11.2 mouse model deletion

Recurrent copy number variations such as the 16p11.2 deletion (del/+) represent a significant genetic risk for neurodevelopmental disorders, showing incomplete penetrance and variable expressivity. Here we describe a striking discordance in a genetic mouse model of human 16p11.2 deletion. Using electrocorticography experiments, we detected that approximately 50% of the isogenic del/+ mice exhibited increased susceptibility to seizures from low doses of the convulsant drug pentylenetetrazole (PTZ). Some of these mice also displayed spontaneous epileptiform brain activity. Next, we used two-photon microscopy tomography (STPT) with c-Fos-GFP (c-Fos+) as a reporter of neuronal activity to obtain images of the entire mouse brain. We examined del/+ mice that exhibited low or high responses to PTZ. Low responders exhibited minimal c-Fos+ activation. However, they had activating circuits that suggest the presence of compensatory mechanisms that suppress seizures. These circuits include the reticular nucleus of the thalamus, the pretectal region, the pedunculopontine nucleus, and the superior colliculus. In contrast, high-responder mice exhibited high c-Fos+ expression, corresponding to a fast seizure phenotype. Using whole-brain imaging, we analyzed brain volume and detected significant differences in the cortex between low- responder and high-responder del/+ mice. In the group with higher seizure susceptibility, we observed a notable decrease in cortical volume with significant differences in several brain regions compared to low responders. Additionally, some high-responder mice showed null expansion areas compared to low-responder mice. During the longitudinal behavioral analysis, the high-responder group of mice exhibited more severe sleep disturbances and an increase in repetitive grooming. Finally, to identify transcriptional correlates of the low- versus high- response phenotypes in adulthood, we performed bulk RNA-seq on somatosensory cortex punches from adult del/+ mice stratified by seizure response. Seizure-prone del/+ mice exhibited reduced activity-dependent immediate early gene signatures and coordinated changes in regulatory programs, including decreased expression of transcriptional regulators and metabotropic glutamate receptor genes, compared to low responders. Together, these findings define two distinct adult del/+ subgroups with convergent circuit, structural, behavioral, and molecular signatures.

neuroscience↗

bioRxiv2026

A consensus atlas of human brain development defines cell type-specific maturation trajectories across the lifespan

Human neurodevelopment is a continuous process that begins prenatally and extends into postnatal life. Current transcriptomic datasets are limited, fragmented across analytical frameworks, precluding comprehensive reconstruction of cellular trajectories linking developmental states to mature cell types. Here we present a consolidated cellular-resolution transcriptomic atlas of human brain development from the onset of neurogenesis to adulthood, covering [~]2.2 million cells from 156 donors across nine studies. All data were reprocessed from raw sequencing reads and annotated within a unified cell-type taxonomy, enabling reliable mapping across the lifespan. Cell types were highly replicable across heterogeneous datasets, enabling us to chart their maturation and cortical layer localization over time. We identify dynamic gene programs predictive of cell-type maturation, validate gene modules tracking known fate transitions, and leverage our atlas scale to characterize rare populations, including microglia. This resource establishes a standardized reference of human brain development and maturation gene modules for future comparisons across model systems, species, and disease states.

neuroscience↗

bioRxiv2026

Monocyte Lineage Expansion Drives Transcriptomic Individuality in Genetically Identical Armadillo Quadruplets

Genetic diversity shapes phenotypes, yet even genetically identical individuals differ. In nine-banded armadillo (Dasypus novemcinctus) quadruplets, we previously showed that allele-specific expression (ASE) imbalances provide a stable molecular fingerprint of individuality. Here, we test whether such transcriptomic individuality reflects functional biological differences. We profiled bulk blood RNA from five cohorts of genetically identical quadruplets across three time points, and found persistent gene-expression signatures that predict individual identity. Focusing on a highly variable cohort, we then performed single-cell RNA-seq and ATAC-seq. In this litter, the most stably distinct individual showed an expanded monocyte-lineage compartment and gene-expression programs enriched for inflammatory and differentiation pathways. These cell-type and regulatory differences were stable over time and robust to experimental leprosy infection. Together, our results link transcriptomic individuality to lasting differences in immune-cell composition, illustrating how early stochastic events can produce persistent, biologically meaningful divergence among genetically identical individuals. TeaserGenetically identical armadillo quadruplets are identifiable by transcriptomic signatures shaped by cellular composition variation.

molecular biology↗

bioRxiv2026

A neofunctionalized flowering antagonist created an evolutionary contingency that channeled Solanaceae adaptation

Neofunctionalization is a rare fate of gene duplication, classically defined as the acquisition of novel functions that potentiate the emergence of new traits. Rather than evolving to function autonomously, neofunctionalized genes may also remain embedded within their ancestral regulatory networks, potentially reshaping the genetic trajectories through which phenotypic change occurs. Testing this hypothesis, we leveraged a pan-genetic platform comprising ten Solanaceae species and show that a paralog of the flowering hormone florigen neofunctionalized into a flowering antagonist and was repeatedly selected during crop domestication and adaptation of wild plants across 50 million years of evolution. Independent selection of cis-regulatory and coding mutations in SELF-PRUNING 5G (SP5G) enabled rapid flowering in the wild ancestor of domesticated tomato from Central America as well as major and indigenous eggplant crop lineages domesticated in Asia and Africa. We further found that cis-regulatory sequence changes reduced SP5G expression and flowering time in wild species native to distinct environments in the Americas and Australia, relationships that we validated by genome editing. Together with similar patterns observed across diverse species and developmental networks, we propose that antagonistic neofunctionalized paralogs create evolutionary contingencies that channel adaptive trajectories across plant lineages.

plant biology↗

bioRxiv2025

XC-ID: De novo identification of the active X chromosome in single-cell RNA-seq

MotivationX chromosome inactivation (XCI) is an epigenetic process that equalizes X-linked gene dosage between females (XX) and males (XY). During early development, one X chromosome in each cell is randomly silenced and clonally inherited, producing a stable mosaic of two epigenetically distinct cell lineages. This mosaicism provides a natural internal control for studying cell-intrinsic regulatory differences between X lineages. However, identifying the active X chromosome in single cells remains difficult due to sparse allelic coverage, dependence on pre-phased references, and biological variability from XCI escape and skew. ResultsWe present XC-ID (X Chromosome inactivation IDentifier), a scalable computational framework for de novo identification of the active X chromosome from single-cell RNA-seq data. XC-ID employs a simulated-annealing algorithm to infer X-linked haplotype structure directly from allelic counts, followed by bootstrap-based confidence estimation to filter uncertain cell assignments. Applied to single-nucleus RNA-seq data from a female Mus musculus hybrid with known genotype, XC-ID achieved >99% accuracy in predicting the active X chromosome. The method remains robust to allelic noise, sequencing errors, and sparsity, and differential expression between inferred X lineages reveals biologically coherent dosage-compensation patterns. AvailabilityXC-ID is available as a Python package with both API and command-line support at https://github.com/jlhjiang/XC-ID.

bioinformatics↗

bioRxiv2025

The misplaced mouse Pax6 interneuron subclass: A cross-species transcriptomic reassignment

Subclass cell-types are strongly conserved across mammalian brains, making the systemic absence of a caudal ganglionic eminence (CGE) Pax6 interneuron subclass in mouse transcriptional atlases unexpected. Through cross-species transcriptomic analysis, we identify a Pax6 subclass homolog in mouse and uncover primate-specific divergence in the Sncg subclass. Our results highlight the pitfalls of relying on single marker genes and provide evolution-aware annotations and marker sets to support robust cross-mammal interneuron comparisons.

neuroscience↗

bioRxiv2025

Epidermal Resident Memory T Cell Fitness Requires Antigen Encounter in the Skin

CD8+ tissue resident memory T cells (TRM) develop from effectors that seed peripheral tissues where they persist providing defense against subsequent challenges. TRM persistence requires autocrine TGF{beta} transactivated by integrins expressed on keratinocytes. TRM precursors that encounter antigen in the epidermis during development outcompete bystander TRM for TGF{beta} resulting in enhanced persistence. ScRNA-seq analysis of epidermal TRM revealed that local antigen experience in the skin resulted in an enhanced differentiation signature in comparison with bystanders. Upon recall, TRM displayed greater proliferation dictated by affinity of antigen experienced during epidermal development. Finally, local antigen experienced TRM differentially expressed TGF{beta}RIII, which increases avidity of the TGF{beta}RI/II receptor complex for TGF{beta}. Selective ablation of Tgfbr3 reduced local antigen experienced TRM capacity to persist, rendering them phenotypically like bystander TRM. Thus, antigen driven TCR signaling in the epidermis during TRM differentiation results in a lower TGF{beta} requirement for persistence and increased proliferative capacity that together enhance epidermal TRM fitness.

immunology↗

bioRxiv2025

Cell Type-Agnostic Transcriptomic Signatures Enable Uniform Comparisons of Neurodevelopment

Single-cell transcriptomics has revolutionized our understanding of neurodevelopmental cell identities, yet, predicting a cell types developmental state from its transcriptome remains a challenge. We perform a meta-analysis of developing human brain datasets comprising over 2.8 million cells, identifying both tissue-level and cell-autonomous predictors of developmental age. While tissue composition predicts age within individual studies, it fails to generalize, whereas specific cell type proportions reliably track developmental time across datasets. Training regularized regression models to infer cell-autonomous maturation, we find that a cell type-agnostic model achieves the highest accuracy (error = 2.6 weeks), robustly capturing developmental dynamics across diverse cell types and datasets. This model generalizes to human neural organoids, accurately predicting normal developmental trajectories (R = 0.91) and disease-induced shifts in vitro. Furthermore, it extends to the developing mouse brain, revealing an accelerated developmental tempo relative to humans. Our work provides a unified framework for comparing neurodevelopment across contexts, model systems, and species.

neuroscience↗

bioRxiv2025

Cluster replicability in single-cell and single-nucleus atlases of the mouse brain

Single-cell RNA sequencing has advanced our understanding of cellular heterogeneity. Ensuring the replicability of identified cell clusters across studies is essential for determining their biological robustness. We assess the replicability of cell clusters identified in two large mouse brain atlases, one generated using single-cell RNA sequencing and the other with single nuclei. Both profile over 4 million cells and group them into over 5000 clusters. Using transcriptome-wide neighbor voting, we identify 2009 reciprocally matched cluster pairs with consistent spatial localization and coordinated gene expression, which were also observed in datasets from multiple species. Reciprocal clusters are enriched in the cerebellum, where lower diversity aids replicability, while the hypothalamuss heterogeneity limits agreement. Distinguishing close clusters is much more challenging than differentiating a cluster from most others, especially when using marker genes. By incorporating replicability data, we provide a stronger foundation for investigating the newly identified clusters and their biological meaning.

neuroscience↗

bioRxiv2024

RNA programmable cell targeting and manipulation with CellREADR

Methods that provide specific, easy, and scalable experimental access to animal cell types and cell states will have broad applications in biology and medicine. CellREADR - Cell access through RNA sensing by Endogenous ADAR (adenosine deaminase acting on RNA), is a programmable RNA sensor-actuator technology that couples the detection of a cell-defining RNA to the translation of an effector protein to monitor and manipulate the cell. The CellREADR RNA device consists of a 5 sensor region complementary to a cellular RNA and a 3 payload coding region; payload translation is gated by the removal of a STOP codon in the sensor region upon base pairing with the cognate cellular RNA through an ADAR-mediated A- to-I editing mechanism ubiquitous to metazoan cells. CellREADR thus highlights the potential for RNA-based monitoring and manipulation of animal cells in ways that are simple, versatile, and generalizable across tissues and species. Here, we describe a detailed protocol for implementing CellREADR experiments in cell cultures and in animals. The procedure includes sensor and payload design, cloning, validation and characterization in mammalian cell cultures. The in vivo animal protocol focuses on AAV-based delivery of CellREADR using brain tissue as examples. We describe current best practices, various experimental controls and trouble-shooting steps. Beginning from sensor design, the validation of RNA sensor-actuators in cell cultures can be completed in 2-3 weeks. The construction of AAV vectors and their in vivo validation and characterization will take additional 6-8 weeks.

molecular biology↗

bioRxiv2024

Cross-expression analysis reveals patterns of coordinated gene expression in spatial transcriptomics

Spatial transcriptomics promises to transform our understanding of tissue biology by molecularly profiling individual cells in situ. A fundamental question they allow us to ask is how nearby cells orchestrate their gene expression. Rather than focus on how these cells (samples) communicate with each other, we reframe the problem to investigate how genes (features) coordinate their expression between neighboring cells. To study these phenomena - called cross-expression - we compare all genes to find pairs that coordinate their expression between adjacent cells, thereby avoiding curating gene lists or annotating cell types. Our end-to-end method recovers ligand-receptor pairs as cross-expressing genes and finds gene combinations that mark anatomical regions, complementing marker gene-based region annotation. Leveraging the overlapping genes across different panels, we use multiple atlas-scale adult mouse brain datasets (~25 million cells, 695 samples, 8 technologies) to create an integrated, meta-analytic cross-expression network, whose communities are enriched in spatial processes such as synaptic signaling and G protein coupled receptor activity. Highlighting cross-expressions biological utility, our network shows that genes Drd1 and Gpr6, which are individually implicated in Parkinsons disease (PD) and are being pursued as therapeutic targets, are cross-expressed within the striatum, hinting at their joint role in PD pathophysiology. We provide an efficient R package (https://github.com/gillislab/CrossExpression/) to computationally analyze and visually explore cross-expression patterns, which allow us to better understand how genes coordinate their expression in space to perform tissue-level functions.

bioinformatics↗

bioRxiv2024

Solanum pan-genomics and pan-genetics reveal paralogs as contingencies in crop engineering

Pan-genomics and genome editing technologies are revolutionizing the breeding of globally cultivated crops. A transformative opportunity lies in the reciprocal exchange of genotype-to-phenotype knowledge of agricultural traits between these major crops and hundreds of locally cultivated indigenous crops, thereby enhancing the diversity and resilience of our food system. However, species-specific genetic variants and their interactions with desired natural or engineered mutations pose barriers to achieving predictable phenotypic effects, even between closely related crops or genotypes. Here, by establishing a pan-genome of the crop-rich genus Solanum and integrating functional genomics and genetics, we show that gene duplication and subsequent paralog diversification are a major obstacle to genotype-phenotype predictability. Despite broad conservation of gene macrosynteny among chromosome-scale references for 22 species, including 13 indigenous crops, hundreds of global and lineage-specific gene duplications exhibited dynamic evolutionary trajectories in paralog sequence, expression, and function, including among members of key domestication gene families. Extending our pan-genome with 10 cultivars of African eggplant and leveraging quantitative genetics and genome editing, we uncovered an intricate history of paralog emergence and evolution within this indigenous crop. The loss of an ancient redundant paralog of the classical regulator of stem cell proliferation and fruit organ number, CLAVATA3 (CLV3), was compensated by a lineage-specific tandem duplication. Subsequent pseudogenization of the derived copy followed by a cultivar-specific structural variant resulted in a single fused functional copy of CLV3 that modifies locule number alongside a newly identified gene controlling the same trait. Our findings demonstrate that paralog diversifications over short evolutionary periods are critical yet underexplored contingencies in trait evolvability and independent crop domestication histories. Unraveling these contingencies is crucial for translating genotype-to-phenotype relationships across related species.

genetics↗

bioRxiv2024

Evaluating Methods for the Prediction of Cell Type-Specific Enhancers in the Mammalian Cortex

Identifying cell type-specific enhancers in the brain is critical to building genetic tools for investigating the mammalian brain. Computational methods for functional enhancer prediction have been proposed and validated in the fruit fly and not yet the mammalian brain. We organized the Brain Initiative Cell Census Network (BICCN) Challenge: Predicting Functional Cell Type-Specific Enhancers from Cross-Species Multi-Omics to assess machine learning and feature-based methods designed to nominate enhancer DNA sequences to target cell types in the mouse cortex. Methods were evaluated based on in vivo validation data from hundreds of cortical cell type-specific enhancers that were previously packaged into individual AAV vectors and retro-orbitally injected into mice. We find that open chromatin was a key predictor of functional enhancers, and sequence models improved prediction of non-functional enhancers that can be deprioritized as opposed to pursued for in vivo testing. Sequence models also identified cell type-specific transcription factor codes that can guide designs of in silico enhancers. This community challenge establishes a benchmark for enhancer prioritization algorithms and reveals computational approaches and molecular information that are crucial for identifying functional enhancers in mammalian cortical cell types. The results of this challenge bring us closer to understanding the complex gene regulatory landscape of the mammalian cortex and to designing more efficient genetic tools to target cortical cell types.

genomics↗

bioRxiv2024

Airway basal stem cells are necessary for the maintenance of functional intraepithelial airway macrophages.

Adult stem cells play a crucial role in tissue homeostasis and repair through multiple mechanisms. In addition to being able to replace aged or damaged cells, stem cells provide signals that contribute to the maintenance and function of neighboring cells. In the lung, airway basal stem cells also produce cytokines and chemokines in response to inhaled irritants, allergens, and pathogens, which affect specific immune cell populations and shape the nature of the immune response. However, direct cell-to-cell signaling through contact between airway basal stem cells and immune cells has not been demonstrated. Recently, a unique population of intraepithelial airway macrophages (IAMs) has been identified in the murine trachea. Here, we demonstrate that IAMs require Notch signaling from airway basal stem cells for maintenance of their differentiated state and function. Furthermore, we demonstrate that Notch signaling between airway basal stem cells and IAMs is required for antigen-induced allergic inflammation only in the trachea where the basal stem cells are located whereas allergic responses in distal lung tissues are preserved consistent with a local circuit linking stem cells to proximate immune cells. Finally, we demonstrate that IAM-like cells are present in human conducting airways and that these cells display Notch activation, mirroring their murine counterparts. Since diverse lung stem cells have recently been identified and localized to specific anatomic niches along the proximodistal axis of the respiratory tree, we hypothesize that the direct functional coupling of local stem cell-mediated regeneration and immune responses permits a compartmentalized inflammatory response.

cell biology↗

bioRxiv2024

Large-scale single-cell profiling of stem cells uncovers redundant regulators of shoot development and yield trait variation

Stem cells in plant shoots are a rare population of cells that produce leaves, fruits and seeds, vital sources for food and bioethanol. Uncovering regulators expressed in these stem cells will inform crop engineering to boost productivity. Single-cell analysis is a powerful tool for identifying regulators expressed in specific groups of cells. However, accessing plant shoot stem cells is challenging. Recent single-cell analyses of plant shoots have not captured these cells, and failed to detect stem cell regulators like CLAVATA3 and WUSCHEL. In this study, we finely dissected stem cell-enriched shoot tissues from both maize and arabidopsis for single-cell RNA-seq profiling. We optimized protocols to efficiently recover thousands of CLAVATA3 and WUSCHEL expressed cells. A cross-species comparison identified conserved stem cell regulators between maize and arabidopsis. We also performed single-cell RNA-seq on maize stem cell overproliferation mutants to find additional candidate regulators. Expression of candidate stem cell genes was validated using spatial transcriptomics, and we functionally confirmed roles in shoot development. These candidates include a family of ribosome-associated RNA-binding proteins, and two families of sugar kinase genes related to hypoxia signaling and cytokinin hormone homeostasis. These large-scale single-cell profiling of stem cells provide a resource for mining stem cell regulators, which show significant association with yield traits. Overall, our discoveries advance the understanding of shoot development and open avenues for manipulating diverse crops to enhance food and energy security.

plant biology↗