bioRxiv Science⌕ Search

Biology subjects

Cui, Y.

Publications and source records attributed to Cui, Y..

At least 109 records · Page 6Linked to original sources

TDP-43 dysregulation of polyadenylation site selection is a defining feature of RNA misprocessing in ALS/FTD and related disorders

Nuclear clearance and cytoplasmic aggregation of the RNA-binding protein TDP-43 are observed in many neurodegenerative disorders, including amyotrophic lateral sclerosis (ALS) and fronto- temporal dementia (FTD). Although TDP-43 dysregulation of splicing has emerged as a key event in these diseases, TDP-43 can also regulate polyadenylation; yet, this has not been adequately studied. Here, we applied the dynamic analysis of polyadenylation from RNA-seq (DaPars) tool to ALS/FTD transcriptome datasets, and report extensive alternative polyadenylation (APA) upon TDP-43 alteration in ALS/FTD cell models and postmortem ALS/FTD neuronal nuclei. Importantly, many identified APA genes highlight pathways implicated in ALS/FTD pathogenesis. To determine the functional significance of APA elicited by TDP-43 nuclear depletion, we examined microtubule affinity regulating kinase 3 (MARK3). Nuclear loss of TDP-43 yielded increased expression of MARK3 transcripts with longer 3UTRs, resulting in greater transcript stability and elevated MARK3 protein levels, which promotes increased neuronal tau S262 phosphorylation. Our findings define changes in polyadenylation site selection as a previously unrecognized feature of TDP-43-driven disease pathology in ALS/FTD and highlight a potentially novel mechanistic link between TDP-43 dysfunction and tau regulation.

neuroscience↗

A conserved Pol II elongator SPT6L mediates Pol V transcription elongation to regulate RNA-directed DNA methylation in Arabidopsis

In plants, the plant-specific RNA polymerase V (Pol V) transcripts non-coding RNAs and provides a docking platform for the association of accessory proteins in the RNA-directed DNA methylation (RdDM) pathway. Various components have been uncovered that are involved in the process of DNA methylation, but it is still not clear how the transcription of Pol V is regulated. Here, we found that the conserved Pol II elongator, SPT6L, bound to thousands of intergenic regions in an RNA polymerase II (Pol II) independent manner. The intergenic enrichment of SPT6L, interestingly, co-occupied with the largest subunit of Pol V (NRPE1) and mutation of SPT6L led to the reduction of DNA methylation but not Pol V enrichment. Furthermore, the association of SPT6L at Pol V loci was dependent on the Pol V associated factor, SPT5L, rather than the presence of Pol V, and the interaction between SPT6L and NRPE1 was compromised in spt5l. Finally, Pol V RIP-seq revealed that SPT6L is required to maintain the amount and length of Pol V transcripts. Our findings thus uncovered the critical role of a Pol II conserved elongator in Pol V mediated DNA methylation and transcription, and shed light on the mutual regulation between Pol V and II in plants.

plant biology↗

Target-aware Molecule Generation for Drug Design Using a Chemical Language Model

Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 7 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC50) of 1.9 M. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field.

biochemistry↗

Dual carbon sequestration with photosynthetic living materials

Natural ecosystems offer efficient pathways for carbon sequestration, serving as a resilient approach to remove CO2 from the atmosphere with minimal environmental impact. However, the control of living systems outside of their native environments is often challenging. Here, we engineered a photosynthetic living material for dual CO2 sequestration by immobilizing photosynthetic microorganisms within a printable polymeric network. The carbon concentrating mechanism of the cyanobacteria enabled accumulation of CO2 within the cell, resulting in biomass production. Additionally, the metabolic production of OH- ions in the surrounding medium created an environment for the formation of insoluble carbonates via microbially-induced calcium carbonate precipitation (MICP). Digital design and fabrication of the living material ensured sufficient access to light and nutrient transport of the encapsulated cyanobacteria, which were essential for long-term viability (more than one year) as well as efficient photosynthesis and carbon sequestration. The photosynthetic living materials sequestered approximately 2.5 mg of CO2 per gram of hydrogel material over 30 days via dual carbon sequestration, with 2.2 {+/-} 0.9 mg stored as insoluble carbonates. Over an extended incubation period of 400 days, the living materials sequestered 26 {+/-} 7 mg of CO2 per gram of hydrogel material in the form of stable minerals. These findings highlight the potential of photosynthetic living materials for scalable carbon sequestration, carbon-neutral infrastructure, and green building materials. The simplicity of maintenance, coupled with its scalability nature, suggests broad applications of photosynthetic living materials as a complementary strategy to mitigate CO2 emissions.

bioengineering↗

STAIG: Spatial Transcriptomics Analysis via Image-Aided Graph Contrastive Learning for Domain Exploration and Alignment-Free Integration

Spatial transcriptomics is an essential application for investigating cellular structures and interactions and requires multimodal information to precisely study spatial domains. Here, we propose STAIG, a novel deep-learning model that integrates gene expression, spatial coordinates, and histological images using graph-contrastive learning coupled with high-performance feature extraction. STAIG can integrate tissue slices without prealignment and remove batch effects. Moreover, it was designed to accept data acquired from various platforms, with or without histological images. By performing extensive benchmarks, we demonstrated the capability of STAIG to recognize spatial regions with high precision and uncover new insights into tumor microenvironments, highlighting its promising potential in deciphering spatial biological intricates.

bioinformatics↗

Variability in single-cell oxygen consumption kinetics

We combined microfabricated devices with multiparameter identification algorithms to probe the variability in size-dependent oxygen consumption parameters of single human hepatic cells. We demonstrate that single cells exhibit an oxygen-dependent metabolic rate, typical of Michaelis-Menten kinetics, and that their maximal oxygen consumption is significantly lower than that of monolayers or 3D hepatic cell aggregates. Notably, we found that clusters of two or more cells competing for a limited oxygen supply reduced their maximal single-cell consumption rate, highlighting their ability to adapt to local resource availability and the presence of nearby cells. Next, we used our high-throughput approach to characterize the covariance of size and oxygen consumption within a cell population. The results show that cooperative behaviour emerges in cell clusters, and that single-cell size and metabolism can be described by a lognormal joint probability density. Our study thus serves as a foundation to connect the metabolic activity of single human hepatocytes to their tissue-or organ-level metabolism as well as describe its size-related variability through scaling laws.

bioengineering↗

Comparative single-cell transcriptomic analysis reveals key differentiation drivers and the potential origin of the vertebrate retina

Despite known single-cell expression profiles in vertebrate retinas, understanding of their developmental and evolutionary expression patterns among homologous cell classes remains limited. We examined and compared approximately 240, 000 retinal cells from four species and found significant similarities among homologous cell classes, indicating inherent regulatory patterns. To understand these shared patterns, we constructed gene regulatory networks for each developmental stage for three of these species. We identified 690 regulons governed by 530 regulators across three species, along with 10 common cell class-specific regulators and 16 preserved regulons. RNA velocity analysis pinpointed conserved putative driver genes and regulators to retinal cell differentiation in both mouse and zebrafish. Investigation of the origins of retinal cells by examining conserved expression patterns between vertebrate retinal cells and invertebrate Ciona intestinalis photoreceptor-related cells implied functional similarities in light transduction mechanisms. Our findings offer insights into the evolutionarily conserved regulatory frameworks and differentiation drivers of vertebrate retinal cells.

bioinformatics↗

HyGAnno: Hybrid graph neural network-based cell type annotation for single-cell ATAC sequencing data

Reliable cell type annotations are crucial for investigating cellular heterogeneity in single-cell omics data. Although various computational approaches have been proposed for single-cell RNA sequencing (scRNA-seq) annotation, high-quality cell labels are still lacking in single-cell ATAC sequencing (scATAC-seq) data, because of extreme sparsity and inconsistent chromatin accessibility between datasets. This calls for novel cell type annotation methods in scATAC-seq, to better explore cell type-specific gene regulatory mechanisms and provide a complementary epigenomic layer to scRNA-seq data. Here, we present a novel automated cell annotation method that transfers cell type information from a well-labeled scRNA-seq reference to an unlabeled scATAC-seq target, via a parallel graph neural network, in a semi-supervised manner. Unlike existing methods that utilize only gene expression or gene activity features, HyGAnno integrates genomewide accessibility peak features to facilitate the training process. In addition, HyGAnno reconstructs a reference-target cell graph that can be used to detect cells with low prediction reliability, according to their specific graph connectivity patterns. HyGAnno was tested using large datasets and demonstrated the advantages of accurate cell annotation, interpretable cell embedding, robustness to noisy reference data, and adaptability to tumor tissues.

bioinformatics↗

FFCM-MRF: An accurate and generalizable cerebrovascular segmentation pipeline for humans and rhesus monkeys based on TOF-MRA

PurposeCerebrovascular segmentation and quantification of vascular morphological features on humans and rhesus monkeys are essential for prevention, diagnosis, and treatment of brain diseases. However, current automated whole-brain vessel segmentation methods are often not generalizable to independent datasets, limiting their usefulness in real-world environments with their heterogeneity in participants, scanners, and species. Materials and MethodsIn this study, we proposed an automated, accurate and generalizable segmentation method for magnetic resonance angiography images called FFCM-MRF. This method integrates fast fuzzy c-means clustering and Markov random field optimization using blood vessel shape priors and spatial constraints. We used a total of 123 human and 44 macaque MRA images scanned at 1.5 T, 3 T, and 7 T MRI from 9 datasets to develop and validate the method. ResultsThe average Dice score coefficients for multiple independent datasets were 69.16-89.63%, with the improvements in FFCM-MRF ranged from 0.16-16.14% compared with state-of-the-art machine learning methods. Quantitative analysis showed that FFCM-MRF can accurately segment major arteries in the Circle of Willis at the base of the brain and smaller distal pial arteries while effectively suppressing noise. Test-retest analysis showed that the model yielded high vascular volume and diameter reliability. ConclusionsOur results demonstrate that the proposed method is highly accurate and reliable and largely independent of variations in field strength, scanner platforms, acquisition parameters, and species. The macaque MRA data and user-friendly open-source toolbox are freely available at OpenNeuro and GitHub to facilitate studies of imaging biomarkers for cerebrovascular and neurodegenerative diseases.

neuroscience↗

Acquisition of Auditory Discrimination Mediated by Different Processes through Two Distinct Circuits Linked to the Lateral Striatum

The striatum, the central hub of cortico-basal ganglia loops, contains functionally heterogeneous subregions distinguished by the topographic patterns of structural connectivity. These subregions mediate various processes of procedural learning. However, it remains unclear when and how striatal subregions engage in the acquisition of sensory stimulus-based decision-making. A neuroimaging of regional brain activity shows that the anterior dorsolateral striatum (aDLS) and posterior ventrolateral striatum (pVLS) in rats are activated in a different temporal pattern during the acquisition phase of auditory discrimination. Chronic and transient pharmacologic manipulations show that the aDLS promotes the behavioral strategy driven by the stimulus- response association while suppressing that by the response-outcome association, and that the pVLS contributes to forming and maintaining the stimulus-response strategy. Electrophysiological recording indicates that subpopulations of aDLS neurons predominantly represent the outcome of specific behaviors at the initial period of discrimination learning, and that pVLS subpopulations encode the beginning and ending of each behavior according to the progress of learning. In addition, other subpopulations of striatal neurons indicate sustained activation after obtaining reward with distinct patterns reflecting the stimulus-response associations. Our findings demonstrate that aDLS and pVLS neurons integrate the new learning of auditory discrimination in spatiotemporally and functionally different manners.

animal behavior and cognition↗

BayesKAT: Bayesian Optimal Kernel-based Test for genetic association studies reveals joint genetic effects in complex diseases

GWAS methods have identified individual SNPs significantly associated with specific phenotypes. Nonetheless, many complex diseases are polygenic and are controlled by multiple genetic variants that are usually non-linearly dependent. These genetic variants are marginally less effective and remain undetected in GWAS analysis. Kernel-based tests (KBT), which evaluate the joint effect of a group of genetic variants, are therefore critical for complex disease analysis. However, choosing different kernel functions in KBT can significantly influence the type I error control and power, and selecting the optimal kernel remains a statistically challenging task. A few existing methods suffer from inflated type 1 errors, limited scalability, inferior power, or issues of ambiguous conclusions. Here, we present a new Bayesian framework, BayesKAT(https://github.com/wangjr03/BayesKAT), which overcomes these kernel specification issues by selecting the optimal composite kernel adaptively from the data while testing genetic associations simultaneously. Furthermore, BayesKAT implements a scalable computational strategy to boost its applicability, especially for high-dimensional cases where other methods become less effective. Based on a series of performance comparisons using both simulated and real large-scale genetics data, BayesKAT outperforms the available methods in detecting complex group-level associations and controlling type I errors simultaneously. Applied on a variety of groups of functionally related genetic variants based on biological pathways, co-expression gene modules, and protein complexes, BayesKAT deciphers the complex genetic basis and provides mechanistic insights into human diseases.

genetics↗

imply: improving cell-type deconvolution accuracy using personalized reference profiles

Real-world clinical samples are often admixtures of signal mosaics from multiple pure cell types. Using computational tools, bulk transcriptomics can be deconvoluted to solve for the abundance of constituent cell types. However, existing deconvolution methods are conditioned on the assumption that the whole study population is served by a single reference panel, which ignores person-to-person heterogeneity. Here we present imply, a novel algorithm to deconvolute cell type proportions using personalized reference panels. imply can borrow information across repeatedly measured samples for each subject, and obtain precise cell type proportion estimations. Simulation studies demonstrate reduced bias in cell type abundance estimation compared with existing methods. Real data analyses on large longitudinal consortia show more realistic deconvolution results that align with biological facts. Our results suggest that disparities in cell type proportions are associated with several disease phenotypes in type 1 diabetes and Parkin-sons disease. Our proposed tool imply is available through the R/Bioconductor package ISLET at https://bioconductor.org/packages/ISLET/.

bioinformatics↗

Enhanced neutralization of SARS-CoV-2 XBB sub-lineages and BA.2.86 by a tetravalent COVID-19 vaccine booster

As the SARS-CoV-2 virus continues to evolve, novel XBB sub-lineages such as XBB.1.5, XBB.1.16, EG.5, HK.3 (FLip), and XBB.2.3, as well as the most recent BA.2.86, have been identified and aroused global concern. Understanding the efficacy of current vaccines and the immune systems response to these emerging variants is critical for global public health. In this study, we evaluated the neutralization activities of sera from participants who received COVID-19 inactivated vaccines, or a booster vaccination of the recently approved tetravalent protein vaccine in China (SCTV01E), or had contracted a breakthrough infection with BA.5/BF.7/XBB virus. Comparative analysis of their neutralization profiles against a broad panel of 30 SARS-CoV-2 sub-lineage viruses revealed that strains such as BQ.1.1, CH.1.1, and all the XBB sub-lineages exhibited heightened resistance to neutralization than previous variants, however, despite the extra mutations carried by emerging XBB sub-lineages and BA.2.86, they did not demonstrate significantly increased resistance to neutralization compared to XBB.1.5. Encouragingly, the SCTV01E booster vaccination consistently induced robust and considerably higher neutralizing titers against all these variants than breakthrough infection did. Cellular immunity assays also showed that the SCTV01E booster vaccination elicited a higher frequency of virus-specific memory B cells but not IFN-{gamma} secreting T cells. Our findings underline the importance of developing novel multivalent vaccines to more effectively combat future viral variants.

immunology↗

Chemogenetic activation of target neurons expressing insect Ionotropic Receptors in the mammalian central nervous system by systemic administration of ligand precursors

The IR-mediated neuronal activation (IRNA) technology allows stimulation of neurons in the brain that heterologously-express members of the insect chemosensory IR repertoire in response to their cognate ligands. In the current protocol, a ligand against the complex consisting of IR84a and IR8a subunits, phenylacetic acid (PhAc), is locally injected into a brain region, because of a low efficiency of PhAc for the delivery into the brain across the blood-brain barrier. To circumvent this invasive injection, here we developed a strategy for activation of target neurons in the brain through peripheral administration with a precursor of PhAc, methyl ester of PhAc (PhAcM), which is efficiently transferred into the brain and converted to the mature ligand by endogenous esterase activities. Peripheral administration of with PhAcM activated IR84a/IR8a-expressing neurons in the locus coeruleus of mice and increased the release of neurotransmitters in their nerve terminal regions. (S)-2-phenylpropionic acid ((S)-PhPr) was newly identified as a ligand for IR84a/IR8a, and peripheral administration with the methyl ester of PhPr with the S-configuration [(S)-PhPrM] caused similar effects on the target neurons. In addition, cell-type specific expression of IR84a/IR8a complex in the striatum of rats was unilaterally induced with a viral vector based on the Cre-loxP system. Peripheral administration with PhAcM or (S)-PhPrM stimulated the neurotransmitter release in the ipsilateral terminal regions of the vector-injected striatum, and PhAcM administration resulted in rotational behavior. Finally, we demonstrated that the metabolites of the peripherally administered-radiolabeled (S)-PhPrM accumulated in the IR84a/IR8a-expressing region in the striatum of the vector-injected rats. These results demonstrate that the systemic IRNA technique provides a powerful strategy for remote manipulation of diverse types of target neurons in the mammalian central nervous system.

neuroscience↗

Shifting Redox Reaction Equilibria on Demand Using an Orthogonal Redox Cofactor

Natural metabolism relies on chemical compartmentalization of two redox cofactors, NAD+ and NADP+, to orchestrate life-essential redox reaction directions. However, in whole cells the reliance on these canonical cofactors limits flexible control of redox reaction direction as these reactions are permanently tied to catabolism or anabolism. In cell-free systems, NADP+ is too expensive in large scale. We have previously reported the use of nicotinamide mononucleotide, (NMN+) as a low-cost, noncanonical redox cofactor capable of specific electron delivery to diverse chemistries. Here, we present Nox Ortho, an NMNH-specific water-forming oxidase, that completes the toolkit to modulate NMNH/NMN+ ratio. This work uncovers an enzyme design principle that succeeds in parallel engineering of six butanediol dehydrogenases as NMN(H)-orthogonal biocatalysts consistently with a 103 - 106 -fold cofactor specificity switch from NAD(P)+ to NMN+. We combine these to produce chiral-pure 2,3-butanediol (Bdo) isomers without interference from NAD(H) or NADP(H) in vitro and in E. coli cells. We establish that NMN(H) can be held at a distinct redox ratio on demand, decoupled from both NAD(H) and NADP(H) redox ratios in vitro and in vivo.

synthetic biology↗

Single-molecule analysis unveils the phosphorylation of FLS2 regulates its spatiotemporal dynamics and immunity

The Arabidopsis thaliana FLAGELLIN-SENSITIVE2 (FLS2), a typical receptor kinase, recognizes the conserved 22 amino acid sequence in the N-terminal region of flagellin (flg22) to initiate plant defense pathways, which was intensively studied in the past decades. However, the dynamic regulation of FLS2 phosphorylation at the plasma membrane after flg22 recognition needs further elucidation. Through single-particle tracking, we demonstrated that upon flg22 treatment, the phosphorylation of Ser-938 in FLS2 impacts its spatiotemporal dynamics and lifetime. Following forster resonance energy transfer-fluorescence lifetime (FRET-FLIM) imaging microscopy and protein proximity indexes (PPI) assays revealed that flg22 treatment increased the co-localization of GFP-tagged FLS2/FLS2S938D but not FLS2S938A with AtRem1.3-mCherry, a sterol-rich lipid marker, indicating that the phosphorylation of FLS2S938 affects FLS2 sorting efficiency to AtRem1.3-associated nanodomains. Importantly, we found that the phosphorylation of Ser-938 enhanced flg22-induced FLS2 internalization and immune responses, demonstrating that the phosphorylation may activate flg22-triggered immunity through partitioning FLS2 into functional AtRem1.3-associated nanodomains, which fills the gap between the FLS2S938 phosphorylation and FLS2-mediated immunity.

cell biology↗

Defense mechanism of a bacterial retron supramolecular assembly

Retrons are a class of multigene antiphage defense system typically consisting of a retron reverse transcriptase, a non-coding RNA, and a cognate effector. Although the triggers for several retron systems have been discovered recently, the full picture of how retron systems sense invading phages and mediate defense remains to be elucidated. Here, we focus on the retron Ec86 defense system and report its modes of activation and action. We identified a phage-encoded DNA cytosine methyltransferase (Dcm) as the trigger of the Ec86 system and show that Ec86 senses msDNA methylation and becomes activated. We further determined the structure of a tripartite retron Ec86 supramolecular assembly, which is primed for activation by Dcm, and demonstrated that the activated system confers defense through depletion of nucleoside derivatives. These findings emphasize the role of retrons being a second line of defense and highlight an emerging theme of anti-phage defense through supramolecular complex assemblies.

microbiology↗

Structure-based self-supervised learning enables ultrafast prediction of stability changes upon mutation at the protein universe scale

Predicting free energy changes ({Delta}{Delta}G) is of paramount significance in advancing our comprehension of protein evolution and holds profound implications for protein engineering and pharmaceutical development. Traditional methods, however, often suffer from limitations such as sluggish computational speed or heavy reliance on biased training datasets. These challenges are magnified when aiming for accurate {Delta}{Delta}G prediction across the vast universe of protein sequences. In this study, we present Pythia, a self-supervised graph neural network tailored for zero-shot {Delta}{Delta}G predictions. In comparative benchmarks with other self-supervised pre-training models and force field-based methods, Pythia outshines its contenders with superior correlations while operating with the fewest parameters, and exhibits a remarkable acceleration in computational speed, up to 105-fold. The efficacy of Pythia is corroborated through its application in predicting thermostable mutations of limonene epoxide hydrolase (LEH) with significant higher experimental success rates. This efficiency propels the exploration of 26 million high-quality protein structures. Such a grand-scale application signifies a leap forward in our capacity to traverse the protein sequence space and potentially enrich our insights into the intricacies of protein genotype-phenotype relationships. We provided a web app at https://pythia.wulab.xyz for users to conveniently execute predictions. Keywords: self-supervised learning, protein mutation prediction, protein thermostability

bioinformatics↗