A5011286905- Single-cell and spatial transcriptomics
- Gene expression and cancer classification
- Cell Image Analysis Techniques
- Extracellular vesicles in disease
- Gene Regulatory Network Analysis
- Advanced Fluorescence Microscopy Techniques
- Wind Energy Research and Development
- MicroRNA in disease regulation
- RNA Research and Splicing
- Bioinformatics and Genomic Networks
- Wind Turbine Control Systems
- Epigenetics and DNA Methylation
- Cancer-related molecular mechanisms research
Central South University
2020-2024
Accurate identification of cell types from single-cell RNA sequencing (scRNA-seq) data plays a critical role in variety scRNA-seq analysis studies. This task corresponds to solving an unsupervised clustering problem, which the similarity measurement between cells affects result significantly. Although many approaches for type have been proposed, accuracy still needs be improved. In this study, we proposed novel framework based on learning, called SSRE. SSRE models relationships subspace...
Abstract With the advent of spatial multi-omics, we can mosaic integrate such datasets with partially overlapping modalities to construct higher dimensional views source tissue. SpaMosaic is a multi-omics integration tool that employs contrastive learning and graph neural networks modality-agnostic batch-corrected latent space suited for analyses like domain identification imputing missing omes. Using simulated experimentally acquired datasets, benchmarked against single-cell methods. The...
Integration of single-cell transcriptome datasets from multiple sources plays an important role in investigating complex biological systems. The key to integration is batch effect removal. Recent methods attempt apply a contrastive learning strategy correct effects. Despite their encouraging performance, the optimal framework for removal still under exploration. We develop improved learning-based correction framework, GLOBE. GLOBE defines adaptive translation transformations each cell...
Integration of growing single-cell RNA sequencing datasets helps better understand cellular identity and function. The major challenge for integration is removing batch effects while preserving biological heterogeneities. Advances in contrastive learning have inspired several learning-based correction methods. However, existing contrastive-learning-based methods exhibit noticeable ad hoc trade-off between mixing preservation heterogeneities (mix-heterogeneity trade-off). Therefore, a...
Abstract Single-cell RNA-sequencing technology (scRNA-seq) brings research to single-cell resolution. However, a major drawback of scRNA-seq is large sparsity, i.e. expressed genes with no reads due technical noise or limited sequence depth during the protocol. This phenomenon also called ‘dropout’ events, which likely affect downstream analyses such as differential expression analysis, clustering and visualization cell subpopulations, cellular trajectory inference, etc. Therefore, there...
scATAC-seq has enabled chromatin accessibility landscape profiling at the single-cell level, providing opportunities for determining cell-type-specific regulation codes. However, high dimension, extreme sparsity, and large scale of data have posed great challenges to cell-type identification. Thus, there been a growing interest in leveraging well-annotated scRNA-seq help annotate data. substantial computational obstacles remain transfer information from scATAC-seq, especially their...
The integration of single-cell multi-omics datasets is critical for deciphering cellular heterogeneities. Mosaic integration, the most general task, poses a greater challenge regarding disparity in modality abundance across datasets. Here, we present ACE, mosaic framework that assembles two types strategies to handle this problem: modality-alignment based strategy (ACE-align) and regression-based (ACE-spec). ACE-align utilizes novel contrastive learning objective explicit alignment uncover...
<title>Abstract</title> With the advent of spatial multi-omics, we can mosaic integrate diverse datasets with partially overlapping modalities to construct consensus multi-modal atlases source tissue. SpaMosaic is a multi-omics integration tool that employs contrastive learning and graph neural networks modality-agnostic batch-corrected latent space suited for analyses like domain identification imputing missing omes. Using simulated experimentally acquired datasets, benchmarked against...
GRN is the core of all living organisms that can explain how genes and their products interact at different levels. To infer potential GRNs from gene expression data remains a great challenge in bioinformatics. Recently, with development single cell RNA sequencing technology, inferring specific involving differentiation or function becomes hot topic. Although there are some methods proposed to accomplish task, it still less than ideal because additional noises pseudo time high dropouts...
Single cell RNA sequencing enables researchers to analyze cellular heterogeneity at high resolution. In the analysis, unsupervised clustering has been a common and powerful way identify types. Nevertheless, dropout rate dimension of scRNA-seq data make it still challenging task. this study, we proposed DeepCI, deep neural network based single method, which simultaneously accomplishes low-dimensional representation learning with implicit imputation data. Tested on real datasets, DeepCI...
Abstract Accurate identification of cell types from single-cell RNA sequencing (scRNA-seq) data plays a critical role in variety scRNA-seq analysis studies. It corresponds to solving an unsupervised clustering problem, which the similarity measurement between cells high dimensional space affects result significantly. Although many approaches have been proposed recently, accuracy type still needs be improved. In this study, we novel framework based on learning, called SSRE. SSRE, model...