Biology subjects

Yin, A.

Publications and source records attributed to Yin, A..

11 recordsLinked to original sources

bioRxiv2026

OCellus: A Language-Model Framework for Single-Cell, Spatial, and Perturbation Biology with Natural-Language Reasoning

Computational modeling of cellular behavior--the virtual cell--has emerged as a stated grand challenge at the intersection of artificial intelligence and biology, yet existing foundation models remain specialized: single-cell models process dissociated transcriptomes only, spatial models require dedicated spatial-aware architectures, and perturbation predictors depend on manually curated knowledge bases that cap generalization. Here we introduce OCellus, a single nine-billion-parameter language model (Qwen3.5-9B) fine-tuned on twenty-two biological tasks that simultaneously addresses all three limitations through three coordinated technical contributions on a shared backbone. First, EvenClock encodes two-dimensional spatial coordinates as eighteen clockface sectors of text, enabling spatial reasoning on a vanilla language model without architectural modification; on ten spatial transcriptomics tasks OCellus attains 77 percent spatial-neighborhood accuracy, 96 percent spatial-cellchat accuracy, and 0.70 proportion-cosine similarity on spatial deconvolution, all without any spatial-aware architectural components. Second, per-gene language-model embeddings replace the Gene Ontology annotations that GEARS depends on, achieving Pearson correlation 0.945 on the Replogle 2022 perturbation benchmark versus 0.84 for GEARS across 457 completely unseen knockout genes. Third, OCellus-Agent provides a Planner-Router-Verifier natural-language interface that achieves 75 percent pipeline accuracy on eighty multi-task queries. Removing language-model embeddings collapses perturbation Pearson to 0.06, confirming that learned functional representations--not graph topology--drive the gain. As a cell-type encoder, OCellus ranks first among fourteen foundation models in linear-probe accuracy at 95.1 percent across four benchmark datasets, and reaches 72.6 percent average across twenty-two evaluated biological tasks--a 57-percentage-point absolute gain over the strongest baseline configuration. As a language model, OCellus uniquely generates natural-language explanations of its predictions, a capability absent from all competing methods. Code, pre-trained model weights, the graph-neural-network module, and the agent system will be made available upon publication.

bioinformatics↗

bioRxiv2026

Mouse models with human antibody repertoires for inducing multiple lineages of HIV-1 broadly neutralizing antibodies

The complementarity determining region (CDR) 1, 2 and 3 of antibodies are the principal antigen contact sites. The heavy chain CDR3 (CDR H3) is highly variable, because it includes random nucleotide additions by the terminal deoxynucleotidyl transferase (TdT). Some broadly neutralizing antibodies (bnAbs) against the human immunodeficiency virus-1 (HIV-1) rely heavily on CDR H3 to recognize conserved epitopes on HIV-1 Envelope (Env) protein. Elicitation of comparable bnAbs is a prime goal of HIV-1 vaccine development, but the shortage of precursor antibodies with suitable CDR H3s in human repertoire is a major challenge for immunogen design. To aid this effort, we generated six mouse models for inducing bnAbs against major HIV-1 Env epitopes. In each mouse model, the immunoglobulin heavy and light chain loci were engineered to predominantly rearrange the germline V, D and J segments of a bnAb lineage. Owing to CDR3 diversity, only a small subset of the V(D)J recombination products may encode variable regions that can engage bnAb epitopes with sufficient affinity to initiate immune response. Therefore, these mouse models can be used to test and optimize immunization strategies to induce bnAbs from rare and diverse precursors in complex antibody repertoires.

immunology↗

bioRxiv2026

Maturation of HIV-1 neutralizing antibodies in a germinal center conditional expression mouse model

In germinal centers, activated B cells modify their antigen receptors through somatic hypermutation (SHM), followed by antigenic selection that favors expansion of high affinity B cells. The affinity maturation process is critical for development of broadly neutralizing antibodies (bnAbs) against the human immunodeficiency virus-1 (HIV-1). BnAbs have been isolated from some people living with HIV-1. Because these antibodies target conserved epitopes of the HIV-1 Envelope (Env) protein, they inhibit a broad spectrum of viruses. Eliciting bnAbs by vaccination is a top priority for HIV-1 prevention, but reproducing the lengthy maturation of bnAbs is a major challenge. The problem is typified by VRC01 class antibodies, which recognize the CD4 binding site of HIV-1 Env protein. To reach the CD4 binding site, antibodies need to navigate through adjacent glycans. Accommodating the glycans requires multiple SHMs in germinal center (GC) B cells, including infrequent events. For this reason, VRC01 vaccine development often stalls at this point. We have generated a mouse model aimed at providing a potential solution for navigating this vaccine design impediment. To this end, we made a mouse model that expresses a stalled VRC01 intermediate conditionally in GC B cells. This system has three advantages: 1) direct expression of the intermediate obviates prior immunization steps, thereby shortening the immunization scheme; 2) the conditional expression system bypasses tolerance control checkpoints that sometimes delete B cells expressing bnAbs; 3) the intermediate responds to immunization in GCs, the physiological site of affinity maturation. With this model, we established an immunization method to mature the VRC01 intermediate into heterologous neutralizing antibodies against viruses with a native glycan shield. Since high mutation load is common among bnAbs, the germinal center conditional expression system could provide a general tool for boost immunogen design to overcome roadblocks in the maturation pathway. Author summaryIn response to antigenic stimulation, cognate B cells become activated and form germinal centers in lymphoid tissues. Germinal center B cells modify their antigen receptors through somatic hypermutation (SHM) of immunoglobulin variable region gene exons, with antigen selecting for high affinity B cells by providing survival advantage. This mechanism accounts for antibody affinity maturaton over the gradual course of an immune response. Affinity maturation is critical for generating potent, neutralizing antibodies against diverse strains of the human immunodeficiency virus-1 (HIV-1). These broadly neutralizing antibodies (bnAbs) are heavily mutated, reflecting lengthy affinity maturation over years of chronic infection. Recapitulating the affinity maturation process is a major challenge for bnAb induction by vaccination. In immunization experiments, bnAb development often stalls at rate limiting steps that involve infrequent, but functionally important, mutational events. Overcoming such obstacles requires boost immunogens that can stimulate the stalled B cells to acquire the requisite mutations. To this end, we recapitulated the maturation arrest of a bnAb lineage by expressing a stalled antibody in mouse germinal center B cells. Using this mouse model, we developed boost immunization conditions that advanced the antibody maturation beyond a roadblock to attain neutralizing activities against heterogenous viruses.

immunology↗

bioRxiv2023

Identifying Naturalistic Movies from Human Brain Activity with High-Density Diffuse Optical Tomography

Modern neuroimaging modalities, particularly functional MRI (fMRI), can decode detailed human experiences. Thousands of viewed images can be identified or classified, and sentences can be reconstructed. Decoding paradigms often leverage encoding models that reduce the stimulus space into a smaller yet generalizable feature set. However, the neuroimaging devices used for detailed decoding are non-portable, like fMRI, or invasive, like electrocorticography, excluding application in naturalistic use. Wearable, non-invasive, but lower-resolution devices such as electroencephalography and functional near-infrared spectroscopy (fNIRS) have been limited to decoding between stimuli used during training. Herein we develop and evaluate model-based decoding with high-density diffuse optical tomography (HD-DOT), a higher-resolution expansion of fNIRS with demonstrated promise as a surrogate for fMRI. Using a motion energy model of visual content, we decoded the identities of novel movie clips outside the training set with accuracy far above chance for single-trial decoding. Decoding was robust to modulations of testing time window, different training and test imaging sessions, hemodynamic contrast, and optode array density. Our results suggest that HD-DOT can translate detailed decoding into naturalistic use.

bioengineering↗

bioRxiv2023

Disease-specific selective vulnerability and neuroimmune pathways in dementia revealed by single cellgenomics

Summary/AbstractThe development of successful therapeutics for dementias requires an understanding of their shared and distinct molecular features in the human brain. We performed single-nuclear RNAseq and ATACseq in Alzheimer disease (AD), Frontotemporal degeneration (FTD), and Progressive Supranuclear Palsy (PSP), analyzing 40 participants, yielding over 1.4M cells from three brain regions ranging in vulnerability and pathological burden. We identify 35 shared disease-associated cell types and 14 that are disease-specific, replicating those previously identified in AD. Disease-specific cell states represent molecular features of disease-specific glial-immune mechanisms and neuronal vulnerability in each disorder, layer 4/5 intra-telencephalic neurons in AD, layer 2/3 intra-telencephalic neurons in FTD, and layer 5/6 near-projection neurons in PSP. We infer intrinsic disease-associated gene regulatory networks, which we empirically validate by chromatin footprinting. We find that causal genetic risk acts in specific neuronal and glial cells that differ across disorders, primarily non-neuronal cells in AD and specific neuronal subtypes in FTD and PSP. These data illustrate the heterogeneous spectrum of glial and neuronal composition and gene expression alterations in different dementias and identify new therapeutic targets by revealing shared and disease-specific cell states.

neuroscience↗

bioRxiv2023

Movement-related modulation in mouse auditory cortex is widespread yet locally diverse

Neurons in the mouse auditory cortex are strongly influenced by behavior, including both suppression and enhancement of sound-evoked responses during movement. The mouse auditory cortex comprises multiple fields with different roles in sound processing and distinct connectivity to movement-related centers of the brain. Here, we asked whether movement-related modulation might differ across auditory cortical fields, thereby contributing to the heterogeneity of movement-related modulation at the single-cell level. We used wide-field calcium imaging to identify distinct cortical fields followed by cellular-resolution two-photon calcium imaging to visualize the activity of layer 2/3 excitatory neurons within each field. We measured each neurons responses to three sound categories (pure tones, chirps, and amplitude modulated white noise) as mice rested and ran on a non-motorized treadmill. We found that individual neurons in each cortical field typically respond to just one sound category. Some neurons are only active during rest and others during locomotion, and those that are responsive across conditions retain their sound-category tuning. The effects of locomotion on sound-evoked responses vary at the single cell level, with both suppression and enhancement of neural responses, and the net modulatory effect of locomotion is largely conserved across cortical fields. Movement-related modulation in auditory cortex also reflects more complex behavioral patterns, including instantaneous running speed and non-locomotor movements such as grooming and postural adjustments, with similar patterns seen across all auditory cortical fields. Our findings underscore the complexity of movement-related modulation throughout the mouse auditory cortex and indicate that movement related modulation is a widespread phenomenon. SIGNIFICANCE STATEMENTThroughout the sensory cortex, neural activity is influenced by behavior. It remains unknown whether primary and higher-order sensory cortical centers are similarly or differentially influenced by movement. We show that movement-related modulation in the mouse auditory cortex is locally complex and heterogeneous, but that at a more macroscopic level, the net effect of movement on primary and higher-order auditory cortex is largely conserved. These data highlight the widespread nature of movement-related modulation and suggest that movement signals may inform neural computations throughout multiple nodes of the sensory cortex.

neuroscience↗

bioRxiv2020

Developmentally Delayed Epigenetic Reprogramming Underlying the Pathogenesis of Preeclampsia

Preeclampsia, a life-threatening pregnancy complication characterized by hypertension and multiorgan damage, affects 2-5% of pregnancies and causes 76,000 deaths per year. Most preeclampsia associated syndromes immediately dispel after removal of placenta, indicating a casual role of placenta in the pathogenesis. Failed transformation of spiral artery due to insufficient invasion and excessive apoptosis of trophoblast suggested developmental defects in preeclampsia placenta. However, the underlying molecular mechanisms that affected placenta development in preeclampsia remained elusive. Here we show that, in preeclampsia placenta, the epigenetic landscape formed during extraembryonic tissue differentiation was disrupted: dramatic chromatin accessibility shift affected known and novel genes implicated in preeclampsia. DNA methylation defects in preeclampsia affected lineage-defining PcG-controlled loci in trophectoderm. LTR12 retrotransposons associated with VCT/SCT-specific genes were hypermethylated. Meanwhile, hundreds of PcG-regulated EVT-specific gene promoters, which otherwise undergone post-ZGA extraembryonic-tissue-specific de novo methylation, were hypomethylated and hyper-activated. Together, these epigenetic defects resulted in placenta developmental delay in preeclampsia. The defective methylation pattern could be detected in serum cfDNA, and could be used to accurately predict preeclampsia at early pregnancy weeks in independent validation cohorts. Our data suggests that the preeclampsia placenta represents a stalled state of epigenetic reprogramming en route of development from trophectoderm to normal placenta.

genomics↗

bioRxiv2019

Interactions of encoding and decoding problems to understand motor control

Learning a map from movement to neural data (Encoding Problem) and vice versa (Decoding Problem) are crucial to understanding motor control. A principled encoding model that understands underlying neural dynamics can help better solve the decoding problem. Here, we develop a new generative encoding model leveraging deep learning that autonomously captures neural dynamics. After training, the model can synthesize spike trains given any observed kinematics, under the guidance of the learned neural dynamics. When neural data from other sessions or subjects are limited, synthesized spike trains can improve cross-session and cross-subject decoding performance of a Brain Computer Interface decoder. For cross-subject, even with ample data for both subjects, neural dynamics learned from a previous subject can transfer useful knowledge that improves the best achievable decoding performance for the new subject. The approach is general and fully data-driven, and hence could apply to neuroscience encoding and decoding problems beyond motor control.

neuroscience↗

bioRxiv2019

Mitochondrial dysfunction and autophagy responses to skeletal muscle stress

Autophagy plays an important role in mitochondrial maintenance, yet many details of skeletal muscle autophagic activity are unresolved in the context of muscle stress and/or damage. Skeletal muscles from mice were stressed either by fatiguing contractions, eccentric contraction-induced injury (ECCI), or freeze injury (FI) to establish a timeline of mitochondrial function and autophagy induction after different forms of muscle stress. Only FI was sufficient to elicit a reduction in mitochondrial function (-88%, p=0.006), yet both ECCI and FI resulted in greater autophagy-related protein content (28-fold, p[≤]0.008) suggesting a tunable autophagic response. Muscles from another cohort of mice were used to determine specific forms of autophagy, i.e., flux and mitochondrial-specific, in response to muscle damage. Mitochondrial-specific autophagy was evident by accumulation of autophagy-related proteins in mitochondrial-enriched muscle fractions following FI (37-fold, p=0.017); however, autophagy flux, assessed by LC3II accumulation with the lysosomal inhibitor chloroquine, was insignificant suggesting a physiological bottleneck in the clearance of dysfunctional organelles following FI. Ulk1 muscle-specific knockout (Ulk1 MKO) mice were used to determine if autophagy is necessary for the recovery of mitochondrial function after muscle damage. Ulk1 MKO mice were weaker (-12%, p=0.012) and demonstrated altered satellite cell dynamics (e.g., proliferation) during muscle regeneration after FI compared to littermate control mice, but determination of autophagy necessity for the recovery of mitochondrial function was inconclusive. This study concludes that autophagy is a tunable cellular response to muscle damaging stress and may influence muscle fiber regeneration through interaction with satellite cells.\n\nKey Points SummaryO_LIMuscle contractility dysfunction is well characterized after many different types of muscle stress however, the timing and magnitude of mitochondrial dysfunction and autophagy induction after different types of muscle stress is largely unknown.\nC_LIO_LIIn this study we found that only traumatic freeze injury causes mitochondria dysfunction compared to fatigue contractions and eccentric contraction-induced injury, and that the autophagic response to muscle stress scales to the magnitude of muscle damage, i.e., freeze vs. eccentric contraction-induced injury.\nC_LIO_LIWe determined that total autophagy-related protein content has a greater response to muscle fiber damage compared to autophagy flux likely reflecting a bottleneck of autophagosomes awaiting degradation following muscle injury.\nC_LIO_LIUsing a skeletal gmuscle-specific autophagy knockout mouse (Ulk1), we found that muscle contractility and satellite cell activity might be influenced by cellular events within the adult muscle fiber following muscle damage.\nC_LI

physiology↗

bioRxiv2019

PGC-1α overexpression partially rescues impaired oxidative and contractile pathophysiology following volumetric muscle loss injury

Volumetric muscle loss (VML) injury is characterized by a non-recoverable loss of muscle fibers due to ablative surgery or severe orthopaedic trauma, that results in chronic functional impairments of the soft tissue. Currently, the effects of VML on the oxidative capacity and adaptability of the remaining injured muscle are unclear. A better understanding of this pathophysiology could significantly shape how VML-injured patients and clinicians approach regenerative medicine and rehabilitation following injury. Herein, the data indicated that VML-injured muscle has diminished mitochondrial content and function (i.e. oxidative capacity), loss of mitochondrial network organization, and attenuated oxidative adaptations to exercise. However, forced PGC-1 over-expression rescued the deficits in oxidative capacity and muscle strength. This implicates physiological activation of PGC1- as a limiting factor in VML-injured muscle adaptive capacity and provides a mechanistic target for regenerative rehabilitation approaches to address the skeletal muscle dysfunction.

physiology↗

bioRxiv2018

Zebrafish as a model to investigate the effects of exercise in cancer

Emerging data indicates that exercise may regulate cancer pathogenesis, but the mechanisms underpinning how it regulates the tumor as well as surrounding microenvironment are poorly understood. Dissecting this complex, highly integrated physiology requires model systems which accurately recapitulate key aspects of human response to exercise, yet permit rapid and unbiased genetic interrogation of relevant pathways. The zebrafish has emerged as a new model for cancer due to its high resolution in vivo imaging and capacity for large-scale, unbiased screening approaches. Here, we have developed a set of tools to study the effects of exercise in a zebrafish model of melanoma. Using a flow chamber, we studied the effects of endurance exercise bouts (3-6 hour/d, 5d/wk for 1 to 3 wks) in both larval and adult zebrafish. The regimen was well tolerated, with no unexpected toxicities or changes in survival. When the zebrafish were transplanted with ZMEL1-GFP melanoma cells, we found that endurance exercise over a 2-week period led to a significant decrease in cancer growth in the larval zebrafish. As zebrafish cancer models show strong conservation in human disease, our findings have direct application to understanding the human exercise/cancer relationship.

cancer biology↗