Biology subjects

Li, I.

Publications and source records attributed to Li, I..

6 recordsLinked to original sources

bioRxiv2025

Modeling patient tissues at molecular resolution with Eva

Tissue structure is essential to function and homeostasis in all organs, and disruptions to structure usually indicate disease. Modeling relationships between structural, molecular, and clinical aspects of tissues could advance new diagnostics and treatment strategies. Although profiling techniques like spatial proteomics can capture these relationships, the data remain challenging to extract insight from. Here, we present Eva, a foundation model for tissue imaging data that learns multi-scale spatial representations of tissues at the molecular, cellular, and sample level. Eva uses a novel vision transformer architecture and is pre-trained on masked reconstruction of over 40 million matched spatial proteomics and histopathology images. We show that Eva excels at a variety of tasks, including cross-modal inference from H&E to proteomics stains, quality control, data annotation, zero-shot retrieval, survival modeling, and patient stratification. Extensive evaluations on held-out validation data demonstrate the versatility and generalizability of the learned embeddings. We anticipate that Eva will accelerate translational science by bridging basic research and clinical practice.

bioinformatics↗

bioRxiv2024

Reactive oxygen species suppress phagocyte surveillance by oxidizing cytoskeletal regulators

Despite their superficial similarities, the phagocytosis of pathogens differs from that of apoptotic cells in their recognition mechanisms and downstream signaling pathways. While the initial stages of these processes have been studied, the cytoskeletal reorganization that follows particle uptake is not well understood. By comparing the uptake of phosphatidylserine (PS)- coated targets versus IgG-opsonized targets of identical size, shape, and rigidity, we noted remarkable differences in the accompanying changes in cell morphology, adhesion and migration that persisted long after phagocytosis. While myeloid cells continued to survey their microenvironment after engulfing PS-coated targets, the uptake of IgG-opsonized targets caused phagocytes to round up, decreased their membrane ruffling, and led to the complete disassembly of podosomes. These changes were associated with increased activation of Rho and a concomitant decrease of Rac activity that collectively resulted in the thickening and compaction of the cortical F-actin cytoskeleton. Rho/formin-induced actin polymers were fastened to the membrane by their preferential interaction with Ezrin-Radixin-Moesin (ERM) proteins, which were necessary for cell compaction and podosome disassembly following ingestion of IgG-coated particles. The source of the distinct responses to PS- versus IgG-targets was the differential activation of the respiratory burst mediated by the NADPH oxidase: reactive oxygen species (ROS), emanating from phagosomes containing IgG-opsonized targets - but not those containing PS-coated ones - directly led to the activation of Rho. Similar findings were made with phagocytes that encountered pathogens or microbial-associated molecular patterns (MAMPS) that instigate the activation of the NADPH oxidase. These results implicate a connection between sensing of harmful particulates, the oxidation of cytoskeletal regulators, and the immune surveillance by myeloid cells that have potentially important consequences for the containment of pathogens.

cell biology↗

bioRxiv2023

The colocatome as a spatial -omic reveals shared microenvironment features between tumour-stroma assembloids and lung cancer specimens

Computational frameworks to quantify and compare microenvironment spatial features of in-vitro patient-derived models and clinical specimens are needed. Here, we acquired and analysed multiplexed immunofluorescence images of human lung adenocarcinoma (LUAD) alongside tumour- stroma assembloids constructed with organoids and fibroblasts harvested from the leading edge (Tumour-Adjacent Fibroblasts;TAFs) or core (Tumour Core Fibroblasts;TCFs) of human LUAD. We introduce the concept of the "colocatome" as a spatial -omic dimension to catalogue all proximate and distant colocalisations between malignant and fibroblast subpopulations in both the assembloids and clinical specimens. The colocatome expands upon the colocalisation quotient (CLQ) through a nomalisation strategy that involves permutation analysis and thereby allows comparisons of CLQs under different conditions. Using colocatome analysis, we report that both TAFs and TCFs protected cancer cells from targeted oncogene treatment by uniquely reorganising the tumour-stroma cytoarchitecture, rather than by promoting cellular heterogeneity or selection. Moreover, we show that the assembloids colocatome recapitulates the tumour-stroma cytoarchitecture defining the tumour microenvironment of LUAD clinical samples and thereby can serve as a functional spatial readout to guide translational discoveries. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=190 SRC="FIGDIR/small/557278v1_ufig1.gif" ALT="Figure 1"> View larger version (60K): org.highwire.dtl.DTLVardef@a79aa3org.highwire.dtl.DTLVardef@199582forg.highwire.dtl.DTLVardef@6c7fb6org.highwire.dtl.DTLVardef@134a2aa_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioengineering↗

bioRxiv2022

Identification of cell types in multiplexed in situ images by combining protein expression and spatial information using CELESTA reveals novel spatial biology

Advances in multiplexed in situ imaging are revealing important insights in spatial biology. However, cell type identification remains a major challenge in imaging analysis, with most existing methods involving substantial manual assessment and subjective decisions for thousands of cells. We propose a novel machine learning algorithm, CELESTA, which uses both cells protein expression and spatial information to identify cell type of individual cells. We demonstrate the performance of CELESTA on multiplexed immunofluorescence in situ images of colorectal cancer and head and neck cancer. Using the cell types identified by CELESTA, we identify tissue architecture associated with lymph node metastasis in HNSCC, which we validate in an independent cohort. By coupling our in situ spatial analysis with single-cell RNA-sequencing data on proximal sections of the same tissue specimens, we identify and validate cell-cell crosstalk associated with lymph node metastasis, demonstrating the power of spatial biology to reveal clinically-relevant cellular interactions.

bioinformatics↗

bioRxiv2021

MaveDB v2: a curated community database with over three million variant effects from multiplexed functional assays

A central problem in genomics is understanding the effect of individual DNA variants. Multiplexed Assays of Variant Effect (MAVEs) can help address this challenge by measuring all possible single nucleotide variant effects in a gene or regulatory sequence simultaneously. Here we describe MaveDB v2, which has become the database of record for MAVEs. MaveDB now contains a large fraction of published studies, comprising over two hundred datasets and three million variant effect measurements. We created tools and APIs to streamline data submission and access, transforming MaveDB into a hub for the analysis and dissemination of these impactful datasets.

genomics↗

bioRxiv2021

Reconstructing co-dependent cellular crosstalk in lung adenocarcinoma using REMI

Cellular crosstalk in tissue microenvironments is fundamental to normal and pathological biological processes. Global assessment of cell-cell interactions (CCI) is not yet technically feasible, but computational efforts to reconstruct these interactions have been proposed. Current computational approaches that identify CCI often make the simplifying assumption that pairwise interactions are independent of one another, which can lead to reduced accuracy. We present REMI (REgularized Microenvironment Interactome), a graph-based algorithm that predicts ligand-receptor (LR) interactions by accounting for LR dependencies on high-dimensional, small sample size datasets. We apply REMI to reconstruct the human lung adenocarcinoma (LUAD) interactome from a bulk flow-sorted RNA-seq dataset, then leverage single-cell transcriptomics data to increase its resolution and identify LR prognostic signatures. We experimentally confirmed colocalization of CTGF:LRP6 as an interaction predicted to be associated with LUAD progression. Our work presents a novel way to reconstruct interactomes and a new approach to identify clinically-relevant cell-cell interactions.

bioinformatics↗