Biology subjects

Gabitto, M. I.

Publications and source records attributed to Gabitto, M. I..

15 recordsLinked to original sources

bioRxiv2026

The SEA-AD DREAM Challenge: Community benchmarking human and AI agent solutions for Alzheimer's disease neuropathology prediction from single-nucleus transcriptomics

Single-nucleus transcriptomic atlases offer an unprecedented opportunity to connect cellular molecular states with Alzheimer's disease (AD) neuropathology, but whether these profiles encode reproducible, predictive information about pathological burden remains unclear. We present the SEA-AD DREAM Challenge, an open, international, model-to-data competition built on the Seattle Alzheimer's Disease Brain Cell Atlas to predict Alzheimer's disease neuropathological severity from single-nucleus RNA-sequencing data. Participants developed containerized models to predict categorical neuropathological staging, including overall Alzheimer's disease neuropathologic change, Braak stage, Thal phase, and CERAD score, as well as quantitative amyloid-{beta} and phospho-tau burden measured by 6E10 and AT8 immunohistochemistry. Across 17 eligible teams from 15 countries, the crowdsourcing framework enabled systematic comparison of diverse computational approaches and surfaced a broad landscape of modeling strategies and candidate predictive features. Top-performing methods achieved near-perfect prediction of categorical staging, with the best submission reaching a quadratic weighted kappa of 1.0 for the Overall AD Neuropathological Change score (ADNC), and competitive prediction of quantitative pathological burden in held-out data, with a best concordance correlation coefficient of 0.48. Post hoc perturbation analyses revealed that top categorical-stage predictions relied heavily on donor-level metadata-driven signals rather than transcriptomic features, whereas quantitative pathology prediction was more robust and supported by transcriptomic and cell-type-associated features with potential biological relevance to AD progression. The challenge also introduced the first AI Agent Track in a DREAM Challenge, providing an early benchmark for autonomous and human-guided agentic model development in single-cell neuroscience. This work demonstrates that single-nucleus transcriptomes encode substantial information about Alzheimer's disease pathology, establishes a reproducible benchmark for molecular neuropathology prediction, and highlights critical principles for designing privacy-preserving, leakage-aware community challenges using deeply phenotyped human brain data.

neuroscience↗

bioRxiv2026

Multiregional single-cell profiling reveals shared and specialized cellular vulnerability in Alzheimer's disease

Alzheimers disease (AD) is defined and staged by the stereotyped, progressive accumulation of amyloid-beta (A{beta}) plaques and hyperphosphorylated tau (pTau) tangles across brain regions. These regions differ substantially in their architecture and function but share largely conserved cellular composition, with some regional specialization. As pathology accumulates, specific neurons are lost and non-neuronal cells shift toward disease-associated states, but whether the cell types affected in any one region are the same across the others has remained unclear. Here we extended the Seattle Alzheimers Disease Brain Cell Atlas (SEA-AD) to ten neo- and allocortical regions spanning the cortical arc of canonical AD staging, profiling approximately seven million nuclei from 84 donors with single-nucleus RNA-seq, ATAC-seq, and Multiome alongside quantitative neuropathology and whole-genome sequencing. Nuclei were mapped to an expanded BRAIN Initiative reference taxonomy of 207 cell types, and a hierarchical pseudo-progression framework derived continuous, donor-level measures of AD pathological burden within each region and across the brain by jointly modeling A{beta} and pTau. Cellular changes were both highly selective and strikingly consistent: only [~]30% of cell types shifted in relative abundance, but those that did changed in a coherent direction across regions. Specific subsets of Sst, Lamp5, Vip, Sncg, and Pvalb inhibitory interneurons and myelinating oligodendrocytes were lost earliest in preclinical donors with minimal pathology, alongside initial emergence of AD-associated microglia; loss of L2/3 and selected deep-layer excitatory types, sharper microglial increases, and reactive astrocyte emergence followed in later-stage donors. Regionally specialized populations were also vulnerable, including expected allocortical types and, unexpectedly, primary visual cortex (V1C)-specialized layer 4 (L4 IT) excitatory neurons and intermixed Sst and Pvalb interneurons. Key changes replicated across three independent cohorts encompassing over 700 additional donors. We examined two complementary vulnerable populations in mechanistic detail: regionally specialized V1C L4 IT neurons lost late despite being widely considered resilient, and pan-cortical Sst interneurons lost earliest in disease. Applying a multi-agentic AI workflow that constructed literature-grounded hypotheses from differential expression to L4 IT neurons nominated hyperexcitability, mediated in part by high NMDA receptor expression, as a convergent vulnerability phenotype. Vulnerable Sst interneurons converged on hyperexcitability through partly distinct pathways, and were enriched for expression of AD GWAS-prioritized genes, linking their vulnerability to the genetic architecture of AD. These data, available at SEA-AD.org, provide a multiregional framework for the community to explore the molecular and cellular changes of AD progression.

neuroscience↗

bioRxiv2026

Multi-Trajectory Pseudotime Inference via Permutation Factorizations

Inferring disease progression from cross-sectional data typically relies on placing individuals along a single latent pseudotemporal axis. However, diseases often exhibit different biomarkers that evolve with different dynamics, making a single shared trajectory insufficient and often leading to biased or inconsistent orderings. We propose a probabilistic framework that maintains a shared global ordering of individuals while allowing clusters of features to follow distinct monotonic trajectories along this ordering. Rather than modeling independent disease progressions, our approach captures heterogeneity through cluster-specific temporal responses defined with respect to a common latent sequence of individuals. The model factorizes inference into a latent permutation over individuals and cluster-specific monotonic functions, enabling flexible yet comparable representations of biomarker dynamics. We jointly infer donor ordering, biomarker clusters, and trajectories within a unified Bayesian framework, using relaxed permutation inference for tractability. Across two major neuropathological datasets in Alzheimers disease, our method recovers biologically meaningful feature clusters and improves alignment with established staging measures compared to single-trajectory baselines. These results show that modeling heterogeneous dynamics relative to a shared progression yields more accurate and interpretable reconstructions of disease progression from cross-sectional data.

neuroscience↗

bioRxiv2026

scVIP: personalized modeling of single-cell transcriptomes for developmental and disease phenotypes

Single-cell transcriptomics resolves cellular heterogeneity within individuals, but connecting molecular states to individual-level phenotypes requires frameworks that explicitly bridge these scales. We present scVIP, a generative model that links gene expression, cell-type composition, and phenotypic measurements within a single probabilistic model, which enables accurate phenotype prediction and interpretable trajectory inference. A cell-type-aware multi-instance learning architecture learns donor embeddings that capture progression while localizing phenotype-associated signals to specific cell populations. Applied across four settings, scVIP accurately predicts cortical developmental age (Pearson r = 0.95), characterizes Huntingtons disease progression (concordance correlation coefficient = 0.90), integrates two Alzheimers disease cohorts recovering disease-relevant microglial and astrocytic programs, and distinguishes healthy from ACPA-positive individuals and non-progressors from early RA individuals, identifying inflammatory T cell programs associated with disease. scVIP enables principled analysis of how cellular states collectively shape organism-level phenotypes across development and disease.

bioinformatics↗

bioRxiv2026

MapMyCells: High-performance mapping of unlabeled cell-by-gene data to reference brain taxonomies

Single-cell mapping methods convert raw, heterogeneous single-cell datasets into interpretable and comparable representations of biological identity. As reference cell-type taxonomies mature, mapping new datasets to shared references has become a central strategy for enabling cross-study integration, reproducible annotation, and cumulative biological knowledge. Here we present MapMyCells, an open-source framework designed to align diverse single-cell omics datasets to hierarchical reference taxonomies with minimal preprocessing. MapMyCells provides out-of-the-box support for an expanding set of high-quality brain cell-type references generated by the Allen Institute for Brain Science, the BRAIN Initiative, and the Seattle Alzheimers Disease Brain Cell Atlas, including whole-brain mouse and human atlases, aging and Alzheimers disease cohorts, and a cross-species consensus taxonomy initially focused on the basal ganglia. MapMyCells enables efficient mapping of hundreds of thousands of cells on standard workstations without specialized hardware, providing a deterministic, scalable, and modality-agnostic approach that is robust across species and molecular assays. The framework produces interpretable confidence metrics and quantitative summaries of mapping performance, allowing users to evaluate assignment precision and accuracy. We demonstrate the mapping of unlabeled transcriptomic, epigenomic, and spatial datasets to reference taxonomies and describe a general workflow for preparing arbitrary hierarchical taxonomies for reference-based mapping. As the ecosystem of single-cell reference atlases expands, MapMyCells offers a practical and reproducible solution for community-scale cell-type annotation and cross-dataset integration, supporting the development of unified and extensible brain cell atlases.

bioinformatics↗

bioRxiv2025

Innovations in spinal cord cell type heterogeneity across vertebrate evolution

Vertebrates display remarkable diversity of sensorimotor behaviors, each adapted to distinct ecological and survival demands. This diversity raises fundamental questions about the evolutionary origin of motor control: do conserved spinal circuits underlie these behaviors, and how have they diverged across species. Recent studies detail spinal cell-type architecture in mammals but comparable, high-resolution atlases of the non-mammalian spinal cord are lacking. Here, we compare spinal cord cell types between fish, frogs, mice and humans, spanning [~]450 million years of evolution. Across species, we define highly conserved programs of cell type specification that segregate spinal neurons into nearly identical cardinal classes during development. This contrasts with adult stages, when spinal cell-type composition selectively diverges for excitatory neuron subpopulations. Using spatial transcriptomics, we localize this species divergence to the superficial, dorsal spinal cord, where variant neuropeptide expression defines mammalian-specific cell types. The most dorsal spinal cord thus emerges as a recently evolved hub for sensory integration in mammals, a neospinal cord analogous to the neocortex.

neuroscience↗

bioRxiv2025

TabVI: Leveraging Lightweight Transformer Architectures to Learn Biologically Meaningful Cellular Representations

Transformer-based foundation models are changing the landscape of natural language processing (NLP), computer vision, and audio, achieving human-level performance across a variety of tasks. Extending these models to single-cell genomics holds significant potential for revealing the cellular and molecular perturbations associated with disease. However, unlike the sequential structure of language, the functional organization of genes is hierarchical and modular. This fundamental difference necessitates the development of meaningful feature selection strategies to adapt NLP transformer architectures effectively. In contrast to many large-scale foundation models, probabilistic models have shown success in learning complex cellular representations from single-cell datasets. In this work, we present TabVI, a probabilistic deep generative model that leverages tabular transformer architectures to improve latent embedding learning. We validate TabVIs performance in cell type annotation and integration benchmarks. We demonstrate that TabVI improves performance across down-stream tasks and is robust to scaling dataset sizes, producing interpretable, sample-specific feature attention masks. TabVI is a lightweight, scientifically-meaningful, transformer architecture for single-cell analysis that excels where large scale foundation models are less effective.

bioinformatics↗

bioRxiv2024

Spinal neuron diversity scales exponentially with swim-to-limb transformation during frog metamorphosis

Vertebrates exhibit a wide range of motor behaviors, ranging from swimming to complex limb-based movements. Here we take advantage of frog metamorphosis, which captures a swim-to-limb-based movement transformation during the development of a single organism, to explore changes in the underlying spinal circuits. We find that the tadpole spinal cord contains small and largely homogeneous populations of motor neurons (MNs) and V1 interneurons (V1s) at early escape swimming stages. These neuronal populations only modestly increase in number and subtype heterogeneity with the emergence of free swimming. In contrast, during frog metamorphosis and the emergence of limb movement, there is a dramatic expansion of MN and V1 interneuron number and transcriptional heterogeneity, culminating in cohorts of neurons that exhibit striking molecular similarity to mammalian motor circuits. CRISPR/Cas9-mediated gene disruption of the limb MN and V1 determinants FoxP1 and Engrailed-1, respectively, results in severe but selective deficits in tail and limb function. Our work thus demonstrates that neural diversity scales exponentially with increasing behavioral complexity and illustrates striking evolutionary conservation in the molecular organization and function of motor circuits across species.

neuroscience↗

bioRxiv2024

The transcriptomic landscape of spinal V1 interneurons reveals a role for En1 in specific elements of motor output

Neural circuits in the spinal cord are composed of diverse sets of interneurons that play crucial roles in shaping motor output. Despite progress in revealing the cellular architecture of the spinal cord, the extent of cell type heterogeneity within interneuron populations remains unclear. Here, we present a single-nucleus transcriptomic atlas of spinal V1 interneurons across postnatal development. We find that the core molecular taxonomy distinguishing neonatal V1 interneurons perdures into adulthood, suggesting conservation of function across development. Moreover, we identify a key role for En1, a transcription factor that marks the V1 population, in specifying one unique subset of V1Pou6f2 interneurons. Loss of En1 selectively disrupts the frequency of rhythmic locomotor output but does not disrupt flexion/extension limb movement. Beyond serving as a molecular resource for this neuronal population, our study highlights how deep neuronal profiling provides an entry point for functional studies of specialized cell types in motor output.

neuroscience↗

bioRxiv2024

Hierarchical Bayesian inference to model continuous phenotypical progression in Alzheimer's Disease

Throughout an organisms life, a multitude of complex and interdependent biological systems transition through biophysical processes that serve as indicators of the underlying biological states. Inferring these latent, unobserved states is a goal of modern biology and neuroscience. However, in many experimental setups, we can at best obtain discrete snapshots of the system at different times and for different individuals. This challenge is particularly relevant in the study of Alzheimers Disease (AD) progression, where we observe the aggregation of pathology in brain donors, but the underlying disease state is unknown. This paper proposes a biophysically motivated Bayesian framework (B-BIND: Biophysical Bayesian Inference for Neurode-generative Dynamics), where the disease state is modeled and continuously inferred from observed quantifications of multiple AD pathological proteins. Inspired by biophysical models, we describe pathological burden as an exponential process. The progression of AD is modeled by assigning a latent score, termed pseudotime, to each pathological state, creating a pseudotemporal order of donors based on their pathological burden. We study the theoretical properties of the model using linearization to reveal convergence and identifiability properties. We provide Markov chain Monte Carlo estimation algorithms, illustrating the effectiveness of our approach with multiple simulation studies across various data conditions. Applying this methodology to data from the Seattle Alzheimers Disease Brain Cell Atlas, we infer the pseudotime ordering of donors. Finally, we analyze the information within each pathological feature to refine the model, focusing on the most informative pathologies. This framework lays the groundwork for continuous pseudotime modeling in the analysis of neurodegenerative diseases.

neuroscience↗

bioRxiv2023

Integrated multimodal cell atlas of Alzheimer's disease

Alzheimers disease (AD) is the most common cause of dementia in older adults. Neuropathological and imaging studies have demonstrated a progressive and stereotyped accumulation of protein aggregates, but the underlying molecular and cellular mechanisms driving AD progression and vulnerable cell populations affected by disease remain coarsely understood. The current study harnesses single cell and spatial genomics tools and knowledge from the BRAIN Initiative Cell Census Network to understand the impact of disease progression on middle temporal gyrus cell types. We used image-based quantitative neuropathology to place 84 donors spanning the spectrum of AD pathology along a continuous disease pseudoprogression score and multiomic technologies to profile single nuclei from each donor, mapping their transcriptomes, epigenomes, and spatial coordinates to a common cell type reference with unprecedented resolution. Pseudo-progression analysis showed two major epochs corresponding with a slow early increase in pathology and a later exponential increase that correlated with cognitive decline. The early phase included inflammatory microglial and reactive astrocyte component, as well as a selective loss of Sst+ inhibitory neuron types in superficial cortical layers, loss of myelinating oligodendrocytes, and up-regulation of a re-myelination program by OPCs. The later phase involved loss of excitatory neurons and Pvalb and Vip neuron subtypes also predominantly in superficial layers. These cell vulnerabilities were also seen in prefrontal cortex and replicated by other independent studies when integrated with the BRAIN Initiative reference. Study data and exploratory tools are freely available to accelerate progress in AD research at SEA-AD.org.

genomics↗

bioRxiv2021

MultiVI: deep generative model for the integration of multi-modal data

Jointly profiling the transcriptional and chromatin accessibility landscapes of single-cells is a powerful technique to characterize cellular populations. Here we present MultiVI, a probabilistic model to analyze such multiomic data and integrate it with single modality datasets. MultiVI creates a joint representation that accurately reflects both chromatin and transcriptional properties of the cells even when one modality is missing. It also imputes missing data, corrects for batch effects and is available in the scvi-tools framework: https://docs.scvi-tools.org/.

bioinformatics↗

bioRxiv2020

CRISPR-Decryptr reveals cis-regulatory elements from noncoding perturbation screens

Clustered Regularly Interspace Short Palindromic Repeats (CRISPR)-Cas9 genome editing methods provide the tools necessary to examine phenotypic impacts of targeted perturbations in high-throughput screens. While these technologies have the potential to reveal functional elements with direct therapeutic applications, statistical techniques to analyze noncoding screen data remain limited. We present CRISPR-Decryptr, a computational tool for the analysis of CRISPR noncoding screens. Our method leverages experimental design: accounting for multiple conditions, controls, and replicates to infer the regulatory landscape of noncoding genomic regions. We validate our method on a variety of mutagenesis, CRISPR activation, and CRISPR interference screens, extracting new insights from previously published data.

bioinformatics↗

bioRxiv2020

A Bayesian nonparametric approach to super-resolution single-molecule localization

We consider the problem of single-molecule identification in super-resolution microscopy. Super-resolution microscopy overcomes the diffraction limit by localizing individual fluorescing molecules in a field of view. This is particularly difficult since each individual molecule appears and disappears randomly across time and because the total number of molecules in the field of view is unknown. Additionally, data sets acquired with super-resolution microscopes can contain a large number of spurious fluorescent fluctuations caused by background noise. To address these problems, we present a Bayesian nonparametric framework capable of identifying individual emitting molecules in super-resolved time series. We tackle the localization problem in the case in which each individual molecule is already localized in space. First, we collapse observations in time and develop a fast algorithm that builds upon the Dirichlet process. Next, we augment the model to account for the temporal aspect of fluorophore photo-physics. Finally, we assess the performance of our methods with ground-truth data sets having known biological structure.

bioinformatics↗

bioRxiv2019

Characterizing the epigenetic landscape of cellular populations from bulk and single-cell ATAC-seq information.

Given its ability to map chromatin accessibility with single base pair resolution, ATAC-seq has become a leading technology to probe the epigenomic landscape of single and aggregated cells. Understanding ATAC-seq data presents distinct analysis challenges, compared to RNA-seq technologies, because of the relative sparseness of the data produced and the interaction of complex noise with multiple chromatin structure scales. Methods commonly used to analyze chromatin accessibility datasets are adapted from algorithms designed to process different experimental technologies, disregarding the statistical and biological differences intrinsic to the ATAC-seq technology. Here, we present a Bayesian statistical approach, termed ChromA, to analyze ATAC-seq data. ChromA annotates the cellular epigenetic landscape by integrating information from replicates, producing a consensus de-noised annotation of chromatin accessibility. Our method can analyze single cell ATAC-seq data, improving cell type identification and correcting many of the biases generated by the sparse sampling inherent in single cell technologies. We validate ChromA on several biological systems, including mouse and human immune cells and find it effective at recovering accessible chromatin, establishing ChromA as a top preforming general platform for mapping the chromatin landscape in different cellular populations from diverse experimental designs.

bioinformatics↗