Biology subjects

Wei, P.

Publications and source records attributed to Wei, P..

At least 37 records · Page 2Linked to original sources

bioRxiv2024

Benchmarking DNA Foundation Models for Genomic Sequence Classification

The rapid evolution of DNA foundation models promises to revolutionize genomics, yet comprehensive evaluations are lacking. Here, we present a comprehensive, unbiased benchmark of five models (DNABERT-2, Nucleotide Transformer V2, HyenaDNA, Caduceus-Ph, and GROVER) across diverse genomic and genetic tasks including sequence classification, gene expression prediction, variant effect quantification, and topologically associating domain (TAD) region recognition, using zero-shot embeddings. Our analysis reveals that mean token embedding consistently and significantly improves sequence classification performance, outperforming other pooling strategies. Model performance varies among tasks and datasets; while general purpose DNA foundation models showed competitive performance in pathogenic variant identification, they were less effective in predicting gene expression and identifying putative causal QTLs compared to specialized models. Our findings offer a framework for model selection, highlighting the impact of architecture, pre-training data, and embedding strategies on performance in genomic and genetic tasks.

genomics↗

bioRxiv2024

A novel statistical framework for meta-analysis of total mediation effect with high dimensional omics mediators in large-scale genomic consortia

Meta-analysis is used to aggregate the effects of interest across multiple studies, while its methodology is largely underexplored in mediation analysis, particularly in estimating the total mediation effect of high-dimensional omics mediators. Large-scale genomic consortia, such as the Trans-Omics for Precision Medicine (TOPMed) program, comprise multiple cohorts with diverse technologies to elucidate the genetic architecture and biological mechanisms underlying complex human traits and diseases. Leveraging the recent established asymptotic standard error of the R-squared (R2)-based mediation effect estimation for high-dimensional omics mediators, we have developed a novel meta-analysis framework requiring only summary statistics and allowing inter-study heterogeneity. Whereas the proposed meta-analysis can uniquely evaluate and account for potential effect heterogeneity across studies due to, for example, varying genomic profiling platforms, our extensive simulations showed that the developed method was more computationally efficient and yielded satisfactory operating characteristics comparable to analysis of the pooled individual-level data when there was no inter-study heterogeneity. We applied the developed method to 8 TOPMed studies with over 5800 participants to estimate the mediation effects of gene expression on age-related variation in systolic blood pressure and sex-related variation in high-density lipoprotein (HDL) cholesterol. The proposed method is available in R package MetaR2M on GitHub.

bioinformatics↗

bioRxiv2024

The subcommissural organ regulates brain development via secreted peptides

The subcommissural organ (SCO) is a gland located at the entrance of the aqueduct of Sylvius in the brain. It exists in species as distantly related as amphioxus and humans, but its function is largely unknown. To explore its function, we compared transcriptomes of SCO and non-SCO brain regions and found three genes, Sspo, Car3, and Spdef, that are highly expressed in the SCO. Mouse strains expressing Cre recombinase from endogenous promoter/enhancer elements of these genes were used to genetically ablate SCO cells during embryonic development, resulting in severe hydrocephalus and defects in neuronal migration and development of neuronal axons and dendrites. Unbiased peptidomic analysis revealed enrichment of three SCO-derived peptides, namely thymosin beta 4, thymosin beta 10, and NP24, and their reintroduction into SCO-ablated brain ventricles substantially rescued developmental defects. Together, these data identify a critical role for the SCO in brain development.

neuroscience↗

bioRxiv2024

Deep learning based genomic breeding of pest-resistant grapevine

Crop pests have profoundly deleterious effects on crop yield and food security. However, conventional pest control depends heavily on the utilization of insecticides, which develops strong pesticide resistance and concerns of food safety. Crop and their wild relatives display diverse levels of pest resistance, indicating the feasibility for breeding of pest-resistant crop varieties. In this study, we integrate deep learning (DL)/machine learning (ML) algorithms, plant phenomics and whole genome sequencing (WGS) data to conduct genomic selection (GS) of pest-resistance in grapevine. We employ deep convolutional neural networks (DCNN) to accurately calculate the severity of damage by pests on grape leaves, which achieves a classification accuracy of 95.3% (Visual Geometry Group 16, VGG16, for binary trait) and a correlation coefficient of 0.94 in regression analysis (DCNN with Pest Damage Score, DCNN-PDS, for continuous trait). We apply DL models to predict and integrate phenotype (both binary and continuous) along with WGS data from 231 grape accessions, conducting Genome-Wide Association Studies (GWAS). This analysis detects a total of 69 QTLs, encompassing 139 candidate genes involved in pathways associated with pest resistance, including jasmonic acid (JA), salicylic acid (SA), ethylene, and other related pathways. Furthermore, through the combination with transcriptome data, we identify specific pest-resistant genes, such as ACA12 and CRK3, which play distinct roles in resisting herbivore attacks. Machine learning-based GS demonstrates a high accuracy (95.7%) and a strong correlation (0.90) in predicting the leaf area damaged by pests as binary and continuous traits in grapevine, respectively. In general, our study highlights the power of DL/ML in plant phenomics and GS, facilitating genomic breeding of pest-resistant grapevine.

plant biology↗

bioRxiv2024

Anti-drift pose tracker (ADPT): A transformer-based network for robust animal pose estimation cross-species

Deep learning-based methods have advanced animal pose estimation, enhancing accuracy and efficiency in quantifying animal behavior. However, these methods frequently experience tracking drift, where noise-induced jumps in body point estimates compromise reliability. Here, we present the Anti-Drift Pose Tracker (ADPT), a transformer-based tool that mitigates tracking drift in behavioral analysis. Extensive experiments across cross-species datasets--including proprietary mouse and monkey recordings and public Drosophila and macaque datasets--demonstrate that ADPT significantly reduces drift and surpasses existing models like DeepLabCut and SLEAP in accuracy. Moreover, ADPT achieved 93.16% identification accuracy for 10 unmarked mice and 90.36% accuracy for freely interacting unmarked mice, which can be further refined to 99.72%, enhancing both anti-drift performance and pose estimation accuracy in social interactions. With its end-to-end design, ADPT is computationally efficient and suitable for real-time analysis, offering a robust solution for reproducible animal behavior studies. The ADPT code is available at https://github.com/tangguoling/ADPT.

neuroscience↗

bioRxiv2023

MiceVAPORDot: A novel automated approach for high-throughput behavioral characterization during E-cigarette exposure in mice

The prosperity of electronic nicotine delivery systems (ENDS) or e-cigarette use has been regarded to lead an increasing risk of nicotine addiction, especially among youth. Understanding and evaluating the behaviors induced by ENDS are fundamental to the study of neuropsychiatric effects of e-cigarettes. However, little is known regarding the behavioral features during e-cigarette exposure in mice. Current behavioral assessments for nicotine addiction are based on nicotine withdrawal-induced anxiety which can be only performed after ENDS exposures. Here we developed MiceVAPORDot, a novel high-throughput tool for automated in situ behavioral characterization during e-cigarette exposure. The integration of a deep learning-based animal pose tracking method by MiceVAPORDot allows precise characterization on behavioral phenotypes of e-vapor exposed mice, which were unable revealed by traditional evaluation methodology such as conditioned place preference and elevated plus maze tests. The behavioral fingerprints recognized by MiceVAPORDot can be used for high-throughput screening on incentive nature of e-cigarette flavors as well as medications for smoking cessation.

animal behavior and cognition↗

bioRxiv2023

Nifuroxazide enhances radiotherapy efficacy against hepatocellular carcinoma by upregulating PD-L1 degradation via the ubiquitin-proteasome pathway

Radiation therapy is a primary treatment for hepatocellular carcinoma (HCC), but its effectiveness can be diminished by various factors. The over-expression of PD-L1 has been identified as a critical reason for radiotherapy resistance. Previous studies have demonstrated that nifuroxazide exerts antitumor activity by damaging the Stat3 pathway, but its efficacy against PD-L1 has remained unclear. In this study, we investigated whether nifuroxazide could enhance the efficacy of radiotherapy in HCC by reducing PD-L1 expression. Our results showed that nifuroxazide significantly increased the sensitivity of tumor cells to radiation therapy by inhibiting cell proliferation and migration while increasing apoptosis in vitro. Additionally, nifuroxazide attenuated the up-regulation of PD-L1 expression induced by irradiation, which may be associated with increased degradation of PD-L1 through the ubiquitination-proteasome pathway. Furthermore, nifuroxazide greatly enhanced the efficacy of radiation therapy in H22-bearing mice by inhibiting tumor growth, improving survival, boosting the activation of T lymphocytes, and decelerating the ratios of Treg cells in spleens. Importantly, nifuroxazide limited the increased expression of PD-L1 in tumor tissues induced by radiation therapy. This study confirms, for the first time, that nifuroxazide can augment PD-L1 degradation to improve the efficacy of radiation therapy in HCC-bearing mice.

immunology↗

bioRxiv2023

MASH: Mediation Analysis of Survival Outcome and High-dimensional Omics Mediators with Application to Complex Diseases

Environmental exposures such as cigarette smoking influence health out-comes through intermediate molecular phenotypes, such as the methylome, transcriptome, and metabolome. Mediation analysis is a useful tool for in-vestigating the role of potentially high-dimensional intermediate phenotypes in the relationship between environmental exposures and health outcomes. However, little work has been done on mediation analysis when the mediators are high-dimensional and the outcome is a survival endpoint, and none of it has provided a robust measure of total mediation effect. To this end, we propose an estimation procedure for Mediation Analysis of Survival outcome and High-dimensional omics mediators (MASH) based on sure independence screening for putative mediator variable selection and a second-moment-based measure of total mediation effect for survival data analogous to the R2 measure in a linear model. Extensive simulations showed good performance of MASH in estimating the total mediation effect and identifying true mediators. By applying MASH to the metabolomics data of 1919 subjects in the Framingham Heart Study, we identified five metabolites as mediators of the effect of cigarette smoking on coronary heart disease risk (total mediation effect, 51.1%) and two metabolites as mediators between smoking and risk of cancer (total mediation effect, 50.7%). Application of MASH to a diffuse large B-cell lymphoma genomics data set identified copy-number variations for eight genes as mediators between the baseline International Prognostic Index score and overall survival.

genomics↗

bioRxiv2023

Spatial IMIX: A Mixture Model Approach to Spatially Correlated Multi-Omics Data Integration

Spatial high-throughput omics data allow scientists to study gene activity in a tissue sample and map where it occurs at the same time. This enables the possibility to investigate important early cancer-initiating events occur in normal-appearing tissue and gene activities that progress and carry through tumor tissue, as defined by "field effect." The "field effect" genes are differentially expressed or methylated genes in the spatially resolved high-dimensional datasets with respect to the pathology subtype in each geographical sample across the tissue region. Current statistical methods for spatially resolved genomics data focus on the association of omics data with spatial coordinates without being able to incorporate and test for the association with the sample subtypes. In addition, analytical methods are underdeveloped for spatially resolved multi-omics data integration. We propose a novel statistical frame-work spatial IMIX to integratively analyze spatially resolved high-dimensional multi-omics data associated with a specific trait, such as sample subtypes while modeling the spatial correlations between samples and the inter-data-type correlations between omics data simultaneously. Through extensive simulations, spatial IMIX demonstrated well-controlled type I error, great power by relaxing the independence assumptions between data types, model selection features, and the ability to control FDR across data types. Data applications to a geographically annotated tissue area of bladder cancer discovered cancer-initiating gene activities and revealed interesting fundamental biological mechanisms through path-way analysis. We have implemented our method in R package spatialimix available at https://github.com/ziqiaow/spatialimix.

genomics↗

bioRxiv2023

Social Behavior Atlas: A computational framework for tracking and mapping 3D close interactions of free-moving animals

The study of social behaviors in animals is essential for understanding their survival and reproductive strategies. However, accurately tracking and analyzing the social interactions of free-moving animals has remained a challenge. Existing multi-animal pose estimation techniques suffer from drawbacks such as the need for extensive manual annotation and difficulty in discriminating between similar-looking animals in close social interactions. In this paper, we present the Social Behavior Atlas (SBeA), a novel computational framework that solves these challenges by employing a deep learning-based video instance segmentation model, 3D pose reconstruction, and unsupervised dynamic behavioral clustering. SBeA framework also involves a multi-camera setup to prevent occlusion, and a novel approach to identify individual animals in close social interactions. We demonstrate the effectiveness of SBeA in tracking and mapping the 3D close interactions of free-moving animals using the example of genetic mutant mice, birds, and dogs. Our results show that SBeA is capable of identifying subtle social interaction abnormalities, and the models and frameworks developed can be applied to a wide range of animal species. SBeA is a powerful tool for researchers in the fields of neuroscience and ecology to study animal social behaviors with a high degree of accuracy and reliability.

neuroscience↗

bioRxiv2023

A cataract-causing Y204X mutation of CRYβB1 promotes C-terminal degradation and higher-order oligomerization

Crystallin (Cry) proteins are a class of main structural proteins of vertebrate eye lens, and their solubility and stability directly determine transparency and refractive power of the lens. Mutation in genes that encode for these Cry proteins is the common cause for congenital cataract. Despite extensive studies, the pathogenic and molecular mechanisms remain unclear. In this study, we identified a novel mutation in CRYBB1 from a congenital cataract family, and demonstrated that this mutation led to an earlier termination of protein translation, resulting in a 49-residue truncation at the CRY{beta}B1 C-terminus. This mutant is susceptible to proteolysis and allows us to determine a 1.2- [A] resolution crystal structure of CRY{beta}B1 without the entire C-terminal domain. In this crystal lattice, two N-terminal domain monomers form a dimer that structurally resembles a wild-type (WT) monomer, but with different surface characteristics. Biochemical analyses suggest that this mutant is significantly more liable to aggregate and degrade, when compared to WT CRY{beta}B1. All our results provide an insight into the mechanism regarding how a mutant Cry contributes to the development of congenital cataract possibly through alteration of inter-protein interactions that result in the opacity of eye lens.

biochemistry↗

bioRxiv2023

Speeding up interval estimation for R2-based mediation effect of high-dimensional mediators via cross-fitting

Mediation analysis is a useful tool in investigating how molecular phenotypes such as gene expression mediate the effect of exposure on health outcomes. However, commonly used mean-based total mediation effect measures may suffer from cancellation of component-wise mediation effects in opposite directions in the presence of high-dimensional omics mediators. To overcome this limitation, we recently proposed a variance-based R-squared total mediation effect measure that relies on the computationally intensive nonparametric bootstrap for confidence interval estimation. In the work described herein, we formulated a more efficient two-stage, cross-fitted estimation procedure for the R2 measure. To avoid potential bias, we performed iterative Sure Independence Screening (iSIS) in two subsamples to exclude the non-mediators, followed by ordinary least squares regressions for the variance estimation. We then constructed confidence intervals based on the newly derived closed-form asymptotic distribution of the R2 measure. Extensive simulation studies demonstrated that this proposed procedure is much more computationally efficient than the resampling-based method, with comparable coverage probability. Furthermore, when applied to the Framingham Heart Study, the proposed method replicated the established finding of gene expression mediating age-related variation in systolic blood pressure and identified the role of gene expression profiles in the relationship between sex and high-density lipoprotein cholesterol level. The proposed estimation procedure is implemented in R package CFR2M.

molecular biology↗

bioRxiv2022

KBeagle: An Adaptive Strategy and Tool for Improvement of Imputation Accuracy and Computing Efficiency

With the development of molecular biology and genetics, deep sequencing technology has become the main way to discover genetic variation and reveal the molecular structure of genome. Due to the complexity of the whole genome segment structure, a large number of missing genotypes have appeared after sequencing, and these missing genotypes can be imputed by genotype imputation method. With the in-depth study of genotype imputation methods, computational intensive and computationally efficient imputation software come into being. Beagle software, as an efficient imputation software, is widely used because of its advantages of low memory consumption, fast running speed and relatively high imputation accuracy. K-Means clustering can divide individuals with similar population structure into a class, so that individuals in the same class can share longer haplotype fragments. Therefore, combining K-Means clustering algorithm with Beagle software can improve the interpolation accuracy. The Beagle and KBeagle method was used to compare the imputation efficiency. The KBeagle method presents a higher imputation matching rate and a shorter computing time. In the genome selection and heritability estimated section, the genotype dataset after imputed, unimputed, and with real genotype show similar prediction accuracy. However the estimated heritability using genotype dataset after imputed is closer to the estimation by the dataset with real genotype. We generated a compounds and efficient imputation method, which presents valuable resource for improvement of imputation accuracy and computing time. We envisage the application of KBeagle will be focus on the livestock sequencing study under strong genetic structure.

bioinformatics↗

bioRxiv2022

Biosynthetic diversification of peptaibol mediates fungus-mycohost interactions

Fungi have evolved a plethora of functionally diverse secondary metabolites (SMs) to enhance their adaptation to various environments. To understand how structurally diverse metabolites contribute to fungal adaptation, we elucidate fungus-mycohost specific interactions mediated by a family of polypeptides, i.e., peptaibols. We specified that peptaibol structural diversification was attributed to the nonspecific substrate recognition by the highly conserved peptaibol synthetases (PSs) in dead wood inhabiting mycoparasitic fungi from the genus Trichoderma. Exemplified by investigation of T. hypoxylon, we characterized a library of 19 amino acid residue peptaibols, named trichohypolins, containing 42 derivatives synthesized by a single PS enzyme (NPS1Th). Elimination of trichohypolin production by the deletion of nps1Th reduced the inhibitory activities of T. hypoxylon on at least 15 saprotrophic host fungi, indicating that peptaibols are essential for interactions of Trichoderma spp. with their mycohosts. Different antagonistic effects of five trichohypolin subfractions SF1-SF5 and two pure compounds trichohypolins A (1) and B (2) on saprotrophic host fungi revealed specific activities of peptaibol derivatives in mediating fungus-mycohost interaction. Our study provides insights into the role of metabolic diversity of biosynthetic pathways in interfungal interactions.

microbiology↗

bioRxiv2022

The effects of electroencephalogram feature-based transcranial alternating current stimulation on working memory and electrophysiology

Transcranial alternating current stimulation (tACS) can influence cognitive functions by modulating brain oscillations. However, results regarding the effectiveness of tACS in regulating cognitive performance have been inconsistent. In the present study, we aimed to find EEG characteristics associated with the improvements in working memory performance, to select tACS stimulus targets and frequency based on this feature, and to explore effects of selected stimulus on verbal working memory. To achieve this goal, we first investigated the EEG characteristics associated with improvements in working memory performance with the aid of EEG analyses and machine learning techniques. These analyses suggested that 8 Hz activity in the prefrontal region was related to accuracy in the verbal working memory task. The tACS stimulus target and pattern were then selected based on the EEG feature. Finally, the selected tACS frequency (8 Hz tACS in the prefrontal region) was applied to modulate working memory. The performance of working memory was improved significantly using the selected stimulation than using 40 Hz and sham stimulation (Especially for participants with low verbal working memory). In conclusion, using EEG features related to positive behavioral changes to select brain regions and stimulation patterns for tACS is an effective intervention for improving working memory. Our results contribute to the groundwork for future tACS closed-loop interventions for cognitive deterioration.

neuroscience↗

bioRxiv2022

Long-term two-photon imaging of spinal cord in freely behaving mice

The spinal cord is critical to the perception of peripheral information under sensory-guided motor behaviors in health and disease. However, the cellular activity underlie spinal cord function in freely behaving animals is not clear. Here, we developed a new method for imaging the spinal cord at cellular and subcellular resolution over weeks under naturalistic conditions. The method involves an improved surgery to reduce spinal movement, and the installation of a miniaturized two-photon microscope to obtain high-resolution imaging in moving mice. In vivo calcium imaging demonstrated that dorsal horn neurons show a sensorimotor program-dependent synchronization and heterogeneity under distinct cutaneous stimuli in behaving mice. The long-term imaging of sensory neurons revealed that in the spinal cord, healthy mice demonstrated stereotyped responses. However, in a neuropathic pain model, plasticity changes and neuronal sensitization were observed. We provide a practical method to study the function of spinal cord on sensory perception and disorders in freely behaving mice.

neuroscience↗

bioRxiv2021

Network design principle for robust oscillatory behaviors with respect to biological noise

Oscillatory behaviors, which are ubiquitous in transcriptional regulatory networks, are often subject to inevitable biological noise. Thus a natural question is how transcriptional regulatory networks can robustly achieve accurate oscillation in the presence of biological noise. Here, we search all two- and three-node transcriptional regulatory network topologies for those robustly capable of accurate oscillation against the parameter variability (extrinsic noise) or stochasticity of chemical reactions (intrinsic noise). We find that, no matter what source of the noise is applied, the topologies containing the repressilator with positive auto-regulation show higher robustness of accurate oscillation than those containing the activator-inhibitor oscillator, and additional positive auto-regulation enhances the robustness against noise. Nevertheless, the attenuation of different sources of noise is governed by distinct mechanisms: the parameter variability is buffered by the long period, while the stochasticity of chemical reactions is filtered by the high amplitude. Furthermore, we analyze the noise of a synthetic human nuclear factor {kappa}B (NF-{kappa}B) signaling network by varying three different topologies, and verify that the addition of a repressilator to the activator-inhibitor oscillator, which leads to the emergence of high-robustness motif--the repressilator with positive auto-regulation, improves the oscillation accuracy in comparison to the topology with only an activator-inhibitor oscillator. These design principles may be applicable to other oscillatory circuits.

systems biology↗

bioRxiv2021

Scaling between DNA and cell size governs bacterial growth homeostasis and resource allocation

Bacteria maintain a stable cell size and a certain DNA content through proliferation as described by classic growth laws. How cells behave when this inherent scaling is broken, however, has rarely been interrogated. Here we engineered Escherichia coli cells with extremely low DNA contents using a tunable synthetic tool CRISPRori that temporarily inhibited chromosome replication initiation. A detailed mechanistic model coupling DNA replication, cell growth, and division revealed a fundamental DNA-centric growth law, which was validated by two observations. First, lineage dynamics were robust to large CRISPRori perturbations with division cycles rapidly restoring through a timer mechanism rather than the adder rule. Second, cellular growth transitioned into a linear regime at low DNA-cytoplasm ratios. Experiments and theory showed that in this regime, cellular resource was redirected to plasmid-borne gene expression. Together with the ability of CRISPRori to bi-directionally modulate plasmid copy numbers, these findings suggest a novel strategy for bio-production enhancement.

systems biology↗