YZYuansheng Zhou

Yuansheng Zhou

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yuansheng Zhou has a total of 11 co-authors. Here are the top co-authors by number of shared publications:

  • Tatyana O. Sharpee

    Salk Institute for Biological Studies

    4 shared publications
  • Guanghua Xiao

    Quantitative Biomedical Research Center, Peter O’Donnell Jr. School of Public Health, University of Texas Southwestern Medical Center, Dallas, USA

    2 shared publications
  • Lin Xu

    Quantitative Biomedical Research Center, Peter O’Donnell Jr. School of Public Health, The University of Texas Southwestern Medical Center, Dallas, USA

    2 shared publications
  • Xue Xiao

    Quantitative Biomedical Research Center, Department of Health Data Science & Biostatistics, Peter O’Donnell Jr. School of Public Health, University of Texas Southwestern Medical Center, Dallas, TX, U.S.A

    2 shared publications
  • Brian H. Smith

    School of Life Sciences, Arizona State University, Tempe, AZ 85287, USA

    1 shared publication
  • Chen Tang

    Quantitative Biomedical Research Center, Department of Health Data Science & Biostatistics, Peter O’Donnell Jr. School of Public Health, University of Texas Southwestern Medical Center, Dallas, TX, U.S.A

    1 shared publication

Latest Publications

  • Open Access

    3D reconstruction of spatial transcriptomics with spatial pattern enhanced graph convolutional neural network

    by Chen Tang, Yuansheng Zhou, Xue Xiao, Lei Dong, Lei Yu, Qiwei Li, Guanghua Xiao, Lin Xu

    Spatially resolved transcriptomics (SRT) is a promising new technology that enables simultaneous analysis of gene expression and spatial information for biomedical research. However, the existing statistical and deep learning algorithms used for analyzing SRT data rely solely on two-dimensional (2D) spatial coordinates, which limits their ability to accurately identify spatial domains, spatially variable genes, cell-to-cell communications, and developmental trajectories in a three-dimensional (3D) spatial manner. To address these limitations, we introduced Spa3D, which utilized the anti-leakage Fourier transform and graph convolutional neural network model to reconstruct 3D-based spatial structures from multiple 2D SRT slices. We demonstrate that Spa3D is appliable to analyze data from various SRT technology platforms and outperforms state-of-art methods by: (I) improving spatial domain identification through 3D reconstruction, (II) elucidating cell-cell communication landscape in the 3D cellular organization, (III) modeling of organ-level tempo-spatial development patterns in a 3D fashion, and (IV) annotating 3D spatial trajectory that are not captured by 2D spatial coordinates. Key points: Most existing spatial omics analysis methods rely on 2D data, limiting their ability to capture full spatial and developmental tissue complexity Spa3D incorporates physical z-axis distances, enabling accurate 3D modeling even when adjacent slices vary in tissue structure and composition Spa3D reconstructs true 3D spatial structures from 2D SRT slices using graph convolutional networks and anti-leakage Fourier transforms Spa3D enhances spatial domain detection, revealing detailed cell-cell communication and organ-level development patterns across multiple spatial omics platforms Spa3D enables novel biological discoveries by revealing spatial features and trajectories not detectable using traditional 2D transcriptomic analysis approaches

  • Top Cited Papers
    Open Access

    Cooperative integration of spatially resolved multi-omics data with COSMOS

    by Yuansheng Zhou, Xue Xiao, Lei Dong, Chen Tang, Guanghua Xiao, Lin Xu

    Recent advancements in biological technologies have enabled the measurement of spatially resolved multi-omics data, yet computational algorithms for this purpose are scarce. Existing tools target either single omics or lack spatial integration. We generate a graph neural network algorithm named COSMOS to address this gap and demonstrated the superior performance of COSMOS in domain segmentation, visualization, and spatiotemporal map for spatially resolved multi-omics data integration tasks.


  • Open Access

    Hyperbolic geometry of gene expression

    by Yuansheng Zhou, Tatyana O. Sharpee


  • Open Access

    Hyperbolic geometry of gene expression

    by Yuansheng Zhou, Tatyana Sharpee

    Understanding the patterns of gene expression is key to elucidating the differences between cell types and across disease conditions. The overwhelmingly large number of genes involved generally makes this problem intractable. Yet, we find that gene expression patterns in five different data datasets can all be described using a small number of variables. These variables describe differences between cells according to a hyperbolic metric. We reach this conclusion by developing methods that, starting with an initial assumption of a Euclidean geometry, can detect the presence of other geometries in the data. The Euclidean metric is used in most of current studies of gene expression, primarily because it is difficult to use other non-linear metrics in high dimensional spaces. The hyperbolic metric is much more suitable for describing data produced by a hierarchically organized network, which is relevant for many biological processes. We find that the hyperbolic effects, but not the space dimensionality, increase with the number of genes that are taken into account. The hyperbolic curvature was the smallest for mouse embryonic stem cells, stronger for mouse kidney, lung and brain cells, and reached the largest value in a set of human cells integrated from multiple sources. We show that taking into account hyperbolic geometry strongly improves the visualization of gene expression data compared to leading visualization methods. These results demonstrate the advantages of knowing the underlying geometry when analyzing high-dimensional data.

  • Open Access

    Hyperbolic geometry of the olfactory space

    by Yuansheng Zhou, Brian H. Smith, Tatyana O. Sharpee

    Natural odors and human perceptual descriptions of odors can be represented within three-dimensional hyperbolic space.


  • Open Access

    Using global t-SNE to preserve inter-cluster data structure

    by Yuansheng Zhou, Tatyana O Sharpee

    The t-distributed Stochastic Neighbor Embedding (t-SNE) method is one of the leading techniques for data visualization and clustering. This method finds lower dimensional embeddings of data points while minimizing distortions in distances between neighboring data points. By construction, t-SNE discards information about large scale structure of the data. We show that adding a global cost function to the t-SNE cost function makes it possible to cluster the data while preserving global inter-cluster data structure. We test the new “global t-SNE” (g-SNE) method on one synthetic and two real data sets on flowers and human brain cells which have significant and meaningful global structures. In all cases, g-SNE outperforms t-SNE in preserving the global structure. The weight parameter λ of the global cost function determines the balance between local and global distances preservations. For the human brain atlas data set, we show the tradeoff of λ in representing global structure of data. Using g-SNE with the optimized λ may therefore yield biological insights into how data is organized on multiple scales. The MATLAB code is available at: https://github.com/gyrheart/gsne