Biology subjects

Ni, K.

Publications and source records attributed to Ni, K..

16 recordsLinked to original sources

bioRxiv2026

Low-Noise Finite-Bandgap FET Biosensors in the Nonlinear Gouy-Chapman-Stern Region: A Universal Calibration Curve

Finite-bandgap charge-based FET biosensors are known to be insensitive to electronic tunneling noise in the semiconductor and interference from weakly-charged proteins in physiological samples. However, their signal is still corrupted by pH fluctuations and condensing counterions in the bulk electrolyte. Here, we demonstrate that electrolyte noise is screened by the Stern layer beyond the linear Debye-Huckel limit, when the interfacial potential drop is larger than the thermal voltage. Using a fully nonlinear Gouy-Chapman-Stern (GCS) theory, a universal calibration curve is derived and shown to collapse the low-noise FET signals for miRNA, extracellular vesicles (EV), and protein, captured either directly or through silica and magnetic nanoparticle charge reporters, over eleven decades of analyte concentration. The theory involves the dissociation constant KD and a gain parameter {chi}, which represents the competing effects of the reporter charge and buffer Debye screening that must be large to produce the low-noise GCS condition. The limit of detection (LOD) is shown to be KD/{chi}, which can be optimized by varying the probe density, charge reporter, and buffer solution. An optimized hybridization buffer then yields {chi} ~ 22 (LOD ~1 pM) for miRNA detection, and multivalent capture reduces KD by four orders of magnitude for EV detection (relative to protein detection) with {chi} ~ 34 (LOD ~10 fM).

biophysics↗

bioRxiv2025

Output-Contingent Working Memory and Decision-Making in Economic Choices

In daily life, economic decisions often unfold sequentially. The brain is thought to compute the subjective value of each option, and comparisons can occur in different reference frames, for example, based on the commodity or presentation order. While primate prefrontal recordings have identified various reference frames for economic choice, it remains unclear how distinct neural mechanisms support them even in similar tasks. To address this, we trained recurrent neural networks (RNNs) on two sequential economic decision-making tasks differing only in output contingencies. Analysis of RNN activity, combined with latent connectivity inference, revealed distinct regimes: commodity-based choices with attractor dynamics and order-based choices with rotational dynamics. Moreover, value and choice representations in the order-based tasks aligned with neural data from a novel experiment where reference frames were not explicitly constrained. Our results suggest that different reference frames emerge depending on task demands and engage distinct working memory and decision-making mechanisms.

neuroscience↗

bioRxiv2025

Distributed Biomarker Discovery for Immuno-Oncology

BackgroundIdentifying robust predictive biomarkers for immuno-oncology (IO) therapy response remains challenging due to the complexity of tumor host interactions and the limited availability of public datasets. While achieving clinical relevance requires biomarkers that generalize across diverse datasets, privacy concerns surrounding sensitive clinical and molecular data further restrict data sharing, impeding progress in biomarker discovery. Materials and MethodsWe developed a scalable, cloud-based distributed pipeline, implemented as an in-silico simulation of federated learning, to enable privacy-preserving integration of clinical and molecular data across institutions. This approach supports the development of genomic prediction algorithms without sharing sensitive patient data. Using curated gene expression profiles, we evaluated the association of tumor microenvironment and IO gene expression signatures with clinical outcomes across pan-cancer, cancer-specific, and treatment-specific settings, and developed a multivariable distributed model to predict IO response. ResultsWe implemented an integrated meta-analysis pipeline using harmonized data across 18 datasets comprising 967 patients and seven cancer types treated with PD-1/PD-L1, CTLA-4, or combination immunotherapies. This approach identified biomarkers specific to lung cancer and CTLA-4 therapy previously unreported in pan-cancer and pan-IO analyses, highlighting the value of context-specific associations. Similar gene expression signatures were associated with progression-free survival and response in both pan-cancer and melanoma datasets, with notable overlap between dual checkpoint blockade and PD-1/PD-L1 therapies. Distributed multivariable XGBoost models outperformed the average of locally trained models, indicating improved generalizability. ConclusionWe present a new distributed pipeline for biomarker discovery in immuno-oncology that preserves patient privacy through federated analysis and secure data handling. This study highlights the importance of larger, diverse datasets to refine cancer- and treatment-specific biomarkers, paving the way for more precise and personalized IO therapies.

bioinformatics↗

bioRxiv2025

AtlasAgent: Vision language model and Agent-guided Framework for Evaluation of Atlas-scale Single-cell Integration

As single-cell omics transitions into the era of AI-virtual cells (AIVC), where large-scale single-cell data integration becomes prevalent, the computational demands of integration evaluation emerge as critical scalability bottlenecks. Traditional integration evaluation pipelines, requiring metrics like k-nearest-neighbor batch effect test (kBET) and Local Inverse Simpsons Index (iLISI) employed by state-of-the-art scIB method, often demand large computational resources and long runtimes, making them infeasible for large scale integration studies. Herein, we present AtlasAgent, the first vision-language model (VLM)-powered and AI agent framework to accelerate atlas-scale integration evaluation at unprecedented speed and scale. We systematically evaluate batch correction quality, biological signal preservation and overcorrection risks using chain-of-thought reasoning in conjunction with few-shot and zero-shot prompting strategies. AtlasAgent completes evaluation within 32 seconds, in contrast to scIBs runtime of 5.55 hours in GPU, while identifying the scIB-determint best integration methods within the top-3 in 88.3% of the time, lowering evaluation time from hours to seconds while preserving alignment with domain expert reasoning. AtlasAgent pioneers the use of VLMs to realize scalable and rapid integration evaluation at atlas scale.

bioinformatics↗

bioRxiv2025

LSH-mediated resolution of R-loops mitigates transcription-replication conflicts to preserve genomic stability in prostate cancer cells

Prostate cancer cells exhibit heightened transcriptional activity to sustain their aggressive phenotype, yet this creates DNA topological stress, leading to RNA: DNA hybrids (R-loops) and DNA damage that compromise genomic stability. LSH, an SNF2-family chromatin remodeler, is frequently dysregulated in cancers and linked to tumor progression, but its mechanistic role in genome maintenance remains unclear. Here, we demonstrate that LSH resolves R-loops in an ATP-dependent manner, mitigating transcription- replication (TR) conflicts in prostate cancer cells. LSH depletion caused R- loop accumulation due to reduced FANCD2 recruitment, exacerbating DNA damage and impairing Rad51 filament formation. Strikingly, LSH deficiency also increased R-loops at promoters of MYC and E2F target genes, stalling RNA polymerase II elongation and disrupting transcriptional programs critical for proliferation. Our findings reveal LSH is a key guardian of genome integrity, resolving R-loops to prevent TR conflicts and maintaining DNA repair fidelity. By linking LSH to R-loop resolution and transcriptional regulation, this study reveals its dual role in prostate cancer pathogenesis-- supporting malignant proliferation while safeguarding genomic stability. These insights nominate LSH as a potential therapeutic target to disrupt oncogenic transcription and induce synthetic lethality in prostate cancer.

cancer biology↗

bioRxiv2025

Epithelial-mesenchymal transition couples with cell cycle arrest at various stages

Numerous computational approaches have been developed to infer cell state transition trajectories from snapshot single-cell data. Most approaches first require projecting high-dimensional data onto a low-dimensional representation; however, this can distort the dynamics of the system. Using epithelial-to-mesenchymal transition (EMT) as a test system, we show that both biology-guided low-dimensional representations and trajectory simulations in high-dimensional state space, not representations obtained with brute force dimensionality-reduction methods, reveal two broad paths of TGF-{beta} induced EMT. The paths arise from the coupling between cell cycle and EMT at either the G1 or G2/M phase, contributing to cell-cycle related EMT heterogeneity. Subsequent multi-plex immunostaining studies confirmed the multiple predicted paths at the protein level. The present study highlights the heterogeneity of EMT paths, emphasizes that caution should be taken when inferring transition dynamics from snapshot single-cell data in two- or three-dimensional representations, and shows that incorporating dynamical information can improve prediction accuracy.

systems biology↗

bioRxiv2025

LivecellX: A Deep-learning-based, Single-Cell Object-Oriented Framework for Quantitative Analysis in Live-Cell Imaging

Live-cell imaging uniquely captures single-cell dynamics in space and time, but robust analysis is limited by segmentation and tracking errors that accumulate across frames. We present LivecellX, a deep-learning-based pipeline that integrates instance-level segmentation error correction with trajectory refinement, leveraging temporal context to recover accurate cell tracks. LivecellX also introduces a benchmark dataset with detailed annotations of common error classes, providing a resource for method development and evaluation. Beyond error correction, the framework incorporates modules for classifying biological processes, reconstructing cell lineages, and analyzing dynamic behaviors. Users can interact with the system programmatically or through a Napari-based graphical interface, enabling flexible integration into diverse workflows. By coupling error-aware correction with comprehensive lineage and dynamics analysis, LivecellX establishes an open, extensible platform that advances the accuracy and scalability of live-cell imaging studies.

biophysics↗

bioRxiv2024

GraphVelo allows inference of multi-modal single cell velocities and molecular mechanisms

RNA velocities and generalizations emerge as powerful approaches for extracting time-resolved information from high-throughput snapshot single-cell data. Yet, several inherent limitations restrict applying the approaches to genes not suitable for RNA velocity inference due to complex transcriptional dynamics, low expression, or lacking splicing dynamics, or data of non-transcriptomic modality. Here, we present GraphVelo, a graph-based machine learning procedure that uses as input the RNA velocities inferred from existing methods and infers velocity vectors lying in the tangent space of the low-dimensional manifold formed by the single cell data. GraphVelo preserves vector magnitude and direction information during transformations across different data representations. Tests on synthetic and experimental single-cell data including viral-host interactome, multi-omics, and spatial genomics datasets demonstrate that GraphVelo, together with downstream generalized dynamo analyses, extends RNA velocities to multi-modal data and reveals quantitative nonlinear regulation relations between genes, virus and host cells, and different layers of gene regulation.

bioinformatics↗

bioRxiv2024

Diverse priming outcomes under conditions of very rare precursor B cells

Rare B cells can have special pathogen-recognition features giving them the potential to make outsized contributions to protective immunity. However, rare naive B cells infrequently participate in immune responses. We investigated how germline-targeting vaccine antigen delivery and adjuvant selection affect priming of exceptionally rare BG18-like HIV broadly neutralizing antibody-precursor B cells (~1 in 50 million) in non-human primates. Only escalating dose (ED) priming immunization using the saponin adjuvant SMNP elicited detectable BG18-like cells in germinal centers (GCs). All groups had strong GC responses, but only ED+SMNP and bolus+SMNP induced BG18-like memory B cells in >50% of animals. One group had vaccine-specific GC responses equivalent to ED+SMNP, but BG18-like memory B cells were rarely detected. Following homologous boosting, BG18-like memory B cells were more frequent in a bolus priming group, but had lower somatic hypermutation and affinities. This outcome was inversely associated with post-prime antibody titers, suggesting antibody feedback can significantly influence rare precursor B cell responses.

immunology↗

bioRxiv2024

Inferring Single-Cell RNA Kinetics from Various Biological Priors

In the context of transcriptional dynamics modeled by ordinary differential equations (ODEs), the RNA level in a single cell is controlled by specific RNA kinetics parameters, which include transcription rate, splicing rate, and degradation rate. Investigating these single-cell RNA kinetics rates is pivotal for understanding RNA metabolism and the heterogeneity of complex tissues. Although metabolic labeling is an effective method to estimate these kinetics rates experimentally, it is not suitable for current large-scale conventional single-cell RNA sequencing (scRNA-seq) data. Moreover, existing methods for scRNA-seq often either neglect certain specific kinetics parameters or use inappropriate ways to fit the parameters. To address these issues, we introduce scRNAkinetics, a parallelized method that fits the kinetics parameters of the ODE for each cell using pseudo-time derived from various biological priors (e.g. cell lineage tree and differentiation potential). This approach allows for the estimation of the relative kinetics of each cell and gene in a scRNA-seq dataset. Validated on simulated datasets, scRNAkinetics can accurately infer the kinetics rates of transcription boosting, multi-branch, and time-dependent RNA degradation systems. Nevertheless, the inferred kinetics trends are concordant with previous studies on metabolic labeling and conventional scRNA-seq datasets. Furthermore, we show that scRNAkinetics can provide valuable insights into different regulatory schemes and validate the coupling between transcription and splicing in RNA metabolism. The open-source implementation of scRNAkinetics is available at https://github.com/poseidonchan/scRNAkinetics.

bioinformatics↗

bioRxiv2023

Robust and Accurate Doublet Detection of Single-Cell Sequencing Data via Maximizing Area Under Precision-Recall Curve

Single-cell sequencing has revolutionized our understanding of cellular heterogeneity by offering detailed profiles of individual cells within diverse specimens. However, due to the limitations of sequencing technology, two or more cells may be captured in the same droplet and share the same barcode. These incidents, termed doublets or multiplets, can lead to artifacts in single-cell data analysis. While explicit experimental design can mitigate these issues with the help of auxiliary cell markers, computationally annotating doublets has a broad impact on analyzing the existing public single-cell data and reduces potential experimental costs. Considering that doublets form only a minor fraction of the total dataset, we argue that current doublet detection methods, primarily focused on optimizing classification accuracy, might be inefficient in performing well on the inherently imbalanced data in the area under the precision-recall curve (AUPRC) metric. To address this, we introduce RADO (Robust and Accurate DOublet detection) - an algorithm designed to annotate doublets by maximizing the AUPRC, effectively tackling the imbalance challenge. Benchmarked on 18 public datasets, RADO outperforms other methods in terms of doublet score and achieves similar performance to the current best methods in doublet calling. Furthermore, beyond its application in single-cell RNA-seq data, we demonstrate RADOs adaptability to single-cell assays for transposase-accessible chromatin sequencing (scATAC-seq) data, where it outperforms other scATAC-seq doublet detection methods. RADOs open-source implementation is available at: https://github.com/poseidonchan/RADO.

bioinformatics↗

bioRxiv2023

Three Conserved Immune Dysfunction and Exclusion Subtypes in Bladder and Pan-cancers: Prognostic and Immunotherapeutic Significance

Molecular subtyping is expected to enable bladder cancer (BC) precise treatment. However, its clinical application remains defective and controversial. Given the significance of tumor immune dysfunction and exclusion (TIDE) in tumor immune escape and immunotherapy, we aimed to develop a novel TIDE-based subtyping method to facilitate personalized management. Transcriptome data of BC was used to evaluate the heterogeneity and the status of TIDE patterns. We identified 69 TIDE biomarker genes and classified BC samples into three subtypes. Subtype I showed the lowest TIDE status and malignancy with the best prognosis and highest sensitivity to immune checkpoint blockade (ICB) treatment, which was enriched of metabolic related signaling pathways. Subtype III represented the highest TIDE status and malignancy with the poorest prognosis and resistance to ICB treatment, resulting from its inhibitory immune microenvironment and T cell terminal exhaustion. Subtype II was in a transitional state with intermediate TIDE level, malignancy, and prognosis. We further confirmed the existence and characteristics of our novel TIDE subtypes using real-world BC samples. This subtyping method was proved to be more efficient than known methods in identifying non-responders to immunotherapy. We also found that combining our TIDE subtypes with known biomarkers can improve the sensitivity and specificity in predicting ICB response. Moreover, besides guiding ICB treatment, this classification approach can assist in selecting the frontline or recommended drugs. Finally, the TIDE subtypes are conserved across pan-tumors. In conclusion, our novel TIDE-based strategy is a powerful clinical tool for BC and pan-cancer patients, and potentially guiding personalized immunotherapy.

bioinformatics↗

bioRxiv2022

Specialized neuronal modules for economic and single-attribute choice in the orbitofrontal cortex

An important component of biological intelligence lies in the capacity to learn and execute various cognitive tasks. This ability may be facilitated by a neural system comprising functionally dissociable neuronal modules. Within the orbitofrontal cortex (OFC) lies a neural circuit that supports economic decision-making across diverse contexts. To investigate the functional specialization of this circuit, we compared the neural activity of OFC neurons in rhesus monkeys during multi-attribute choice (MC) and single-attribute choice (SC) tasks. In MC, the monkeys made subjective tradeoffs between competing attributes, whereas in SC, choices were deterministic based on a single attributes value. Neuronally, in MC, OFC neurons primarily encoded offer value, chosen value, and choice in goods space. Upon transitioning from MC to SC, a significant portion of MC-modulated neurons disengaged, while a separate set of neurons emerged to encode SC-related variables in a spatial reference frame. Notably, we observed the clustering of value-encoding neurons in MC but not SC. In essence, our findings suggest that choice tasks involving different mental processes are represented by dissociable neuronal modules within the OFC.

neuroscience↗

bioRxiv2021

Translation rescue by targeting Ppp1r15a upstream open reading frame in vivo

The eIF2 initiation complex is central to maintaining a functional translation machinery. Extreme stress such as life-threatening sepsis exposes vulnerabilities in this tightly regulated system, resulting in an imbalance between the opposing actions of kinases and phosphatases on the main regulatory subunit eIF2. Here, we report that translation shutdown is a hallmark of established sepsis-induced kidney injury brought about by excessive eIF2 phosphorylation and sustained by blunted expression of the counterregulatory phosphatase subunit Ppp1r15a. We determined that the blunted Ppp1r15a expression persists because of the presence of an upstream open reading frame (uORF). Overcoming this barrier with genetic approaches enabled the derepression of Ppp1r15a, salvaged translation and improved kidney function in an endotoxemia model. We also found that the loss of this uORF has broad effects on the composition and phosphorylation status of the immunopeptidome that extended beyond the eIF2 axis. Collectively, our findings define the breath and potency of the highly conserved Ppp1r15a uORF and provide a paradigm for the design of uORF-based translation rheostat strategies. The ability to accurately control the dynamics of translation during sepsis will open new paths for the development of therapies at codon level precision.

physiology↗

bioRxiv2021

Transiently increased intercommunity regulation characterizes concerted cell phenotypic transition

Phenotype transition takes place in many biological processes such as differentiation and reprogramming. A fundamental question is how cells coordinate switching of expressions of clusters of genes. Through analyzing single cell RNA sequencing data in the framework of transition path theory, we studied how such a genome-wide expression program switching proceeds in five different cell transition processes. For each process we reconstructed a reaction coordinate describing the transition progression, and inferred the gene regulation network (GRN) along the reaction coordinate. In all processes we observed common pattern that the overall effective number and strength of regulation between different communities increase first and then decrease. The change accompanies with similar change of the GRN frustration, defined as overall conflict between the regulation received by genes and their expression states, and GRN heterogeneity. While studies suggest that biological networks are modularized to contain perturbation effects locally, our analyses reveal a general principle that during a cell phenotypic transition, intercommunity interactions increase to concertedly coordinate global gene expression reprogramming, and canalize to specific cell phenotype as Waddington visioned.

systems biology↗

bioRxiv2021

Assembly status transition offers an avenue for activity modulation of a supramolecular enzyme

Nature has evolved many supramolecular proteins assembled in certain, sometimes even seemingly oversophisticated, morphological manners. The rationale behind such evolutionary efforts is often poorly understood. Here we provide atomic-resolution insights into how the dynamic building of a structurally complex enzyme with higher-order symmetry offers amenability to intricate allosteric regulation. We have established the functional coupling between enzymatic activity and protein morphological states of glutamine synthetase (GS), an old multi-subunit enzyme essential for cellular nitrogen metabolism. Cryo-EM structure determination of GS in both the catalytically active and inactive assembly states allows us to reveal an unanticipated self-assembly-induced dynamics-driven allosteric paradigm, in which the remote interactions between two subcomplex entities significantly rigidify the otherwise structurally fluctuating active sites, thereby regulating activity. We further show in vivo evidences that how the enzyme morphology transitions could be modulated by cellular factors on demand. Collectively, our data present an example of how assembly status transition offers an avenue for allosteric modulation, and sharpens our mechanistic understanding of allostery, dynamics, cooperativity, and other complex functional and regulatory properties of supramolecular enzymes.

biochemistry↗