Abstract:Abstract Spatial transcriptomics (ST) data holds enormous potential to provide new perspectives for biologists seeking to understand the tumor tissue microenvironments. Cells in the same microenvironment are always exposed to biochemical conditions displacing specific spatial patterns. To truly dismantle the tissue microenvironment heterogeneity as well as cells’ diverse responses to the variable microenvironment, we rely on spatial information to dissect the tissue into different functional regions, which are spatially dependent and biologically meaningful. We have developed a statistical method, smooth Low Rank approximation and Clustering (smoothLRC) to model the spatial transcriptomic data as its low-rank approximation with spatial smoothness. smoothLRC accomplishes this by extracting the latent low-rank structure of the expression profiles. We present a regularized maximum likelihood estimator approximating the noisy observed expression matrix. This estimator incorporates spatial information and addresses expression dropout. It uses a low-rank Poisson distribution to account for the noise that plagues ST-based gene expression data with severe dropout events and a spatial smoothness penalty to encourage neighboring cells to fall into the same functional region. A dropout is an event where genes expressed in a given cell or region are incorrectly measured as unexpressed. SmoothLRC has the following key features: 1) it is a statistically rigorous method; 2) the spatially smooth low-rank subspace could enable meaningful functional dissection, which is an unprecedented capability; 3) the low-rank structures could be used to recover the original expression, saving many expression values that are not captured. Citation Format: Xinyu Zhou, Alexander White, Tingbo Guo, Pengtao Dang, Yuhui Wei, Xiao Wang, Kaman So, Chi Zhang, Sha Cao. Functional dissection of complex tissue microenvironment using spatial transcriptomics data. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4274.

Smoother: A Unified and Modular Framework for Incorporating Structural Dependency in Spatial Omics Data

Smoother: on-the-fly processing of interactome data using prefix sums

MorphLink: Bridging Cell Morphological Behaviors and Molecular Dynamics in Multi-modal Spatial Omics

SMDB: a Spatial Multimodal Data Browser

Abstract 4274: Functional dissection of complex tissue microenvironment using spatial transcriptomics data

Joint dimension reduction and clustering analysis of single-cell RNA-seq and spatial transcriptomics data

CellsFromSpace: A fast, accurate and reference-free tool to deconvolve and annotate spatially distributed Omics data

A robust statistical approach for finding informative spatially associated pathways

Spatial Reconstruction of Oligo and Single Cells by De Novo Coalescent Embedding of Transcriptomic Networks

Mapping Cellular Interactions from Spatially Resolved Transcriptomics Data

SpatialCells: automated profiling of tumor microenvironments with spatially resolved multiplexed single-cell data

Methods for Joint Imaging and RNA-seq Data Analysis

An integrative spatial multi-omic workflow for unified analysis of tumor tissue

Automated spatial omics landscape analysis approach reveals novel tissue architectures in ulcerative colitis

Tissue characterization at an enhanced resolution across spatial omics platforms with deep generative model

scHolography: a computational method for single-cell spatial neighborhood reconstruction and analysis

A Unified Probabilistic Framework for Modeling and Inferring Spatial Transcriptomic Data

Discovery and generalization of tissue structures from spatial omics data

SIMVI reveals intrinsic and spatial-induced states in spatial omics data

Learning tissue representation by identification of persistent local patterns in spatial omics data