<h1>Hepatotoxicity Assessment of Microplastics in Huh7 and HepG2 Cells through High-Content Analysis</h1> <p>This repository contains the following resources produced during the Master’s research of <strong>Mariana Rodrigues da Silva</strong> at the <em>NanoCell Interactions Lab</em>, University of Campinas (UNICAMP), which focused on investigating the <strong>in vitro hepatotoxicity</strong> of microplastics (PS-MPs) using <strong>HepG2</strong> and <strong>Huh7</strong> cell lines, image-based assays (<strong>High-Content Analysis</strong>), and cellular profiling.</p> <p>The four main folders are organized as follows:</p> <pre> ├── Cellpose_Models │ ├── Calcein-PI-Hoechst │ ├── Total-Association-of-PS-MPs │ ├── CellPainting │ └── LiveCellPainting ├── CellProfiler_Pipelines │ ├── Calcein-PI-Hoechst │ ├── Total-Association-of-PS-MPs │ ├── LiveCellPainting │ └── CellPainting ├── Images │ ├── Calcein-PI-Hoechst │ ├── Total-Association-of-PS-MPs │ ├── LiveCellPainting │ └── CellPainting └── Jupyter_Notebooks </pre> <h2>Description of Each Folder:</h2> <h3><strong>Cellpose_Models</strong>:</h3> <p>Contains trained models using <strong>Cellpose (version 2.2.3)</strong> for segmenting cellular structures (cells and/or nuclei) in Huh7 and HepG2 cell lines. Each subfolder corresponds to a specific imaging assay.</p> <h3><strong>CellProfiler_Pipelines</strong>:</h3> <p>Contains pipelines developed in <strong>CellProfiler (version 4.2.5)</strong> to extract cellular features from Huh7 and HepG2 cell lines. Each subfolder corresponds to a specific imaging assay.</p> <h3><strong>Images</strong>:</h3> <p>Contains raw image datasets for each imaging assay, organized into subfolders by assay name. Each subfolder contains two subfolders corresponding to the cell type (<strong>Huh7</strong> or <strong>HepG2</strong>; except for <code>CellPainting</code>, which only contains Huh7). These folders are further subdivided into three independent replicate folders, each containing a subfolder named <code>platemap</code> with the plate layout for the 96-well plate used during image acquisition. The metadata specifies the compounds and concentrations used to treat each well.</p> <p>The cells were stained using six imaging panels:</p> <ol> <li><strong>Calcein-PI-Hoechst</strong> (cell viability assay): <ul> <li>Images acquired with a <strong>10× objective</strong> on the Cytation 5 using the following LED cube channels:</li> <li><strong>DAPI</strong>: nuclei of total cells (excitation/emission = 377/447 nm)</li> <li><strong>GFP</strong>: live cells (excitation/emission = 445/510 nm)</li> <li><strong>PI</strong>: dead cells (excitation/emission = 531/647 nm)</li> </ul> </li> <li><strong>Total Association of PS-MPs with Cells Over Time</strong> (interaction assay between cells and fluorescent PS-MPs over time): <ul> <li>Images acquired with a <strong>20× objective</strong> on the Cytation 5 using the following LED cube channels:</li> <li><strong>DAPI</strong>: nuclei (excitation/emission = 377/447 nm)</li> <li><strong>GFP</strong>: live cells (excitation/emission = 445/510 nm)</li> <li><strong>PI</strong>: PS-MPs (excitation/emission = 531/647 nm)</li> </ul> </li> <li><strong>Live Cell Painting</strong> (cell profiling assay, Garcia-Fossa et al., 2024; <a href="https://doi.org/10.1101/2024.08.28.610144">DOI</a>): <ul> <li>Images acquired with a <strong>20× objective</strong> on the Cytation 5 using the following LED cube channels:</li> <li><strong>DAPI</strong>: nuclei (excitation/emission = 377/447 nm)</li> <li><strong>GFP</strong>: nuclei, nucleoli, cytoplasm (excitation/emission = 445/510 nm)</li> <li><strong>PI</strong>: acidic vesicles (excitation/emission = 531/647 nm)</li> </ul> </li> <li><strong>Cell Painting</strong> (cell profiling assay, Bray et al., 2016; <a href="https://doi.org/10.1038/nprot.2016.105">DOI</a>): <ul> <li>Images acquired with a <strong>20× objective</strong> on the Cytation 5 using the following LED cube channels:</li> <li><strong>DAPI</strong>: plasma membrane and nuclei (excitation/emission = 377/447 nm)</li> <li><strong>GFP</strong>: endoplasmic reticulum (excitation/emission = 445/510 nm)</li> <li><strong>PI</strong>: actin cytoskeleton and Golgi complex (excitation/emission = 531/647 nm)</li> <li><strong>CY5</strong>: mitochondria (excitation/emission = 628/685 nm)</li> </ul> </li> </ol> <h3><strong>Jupyter_Notebooks</strong>:</h3> <p>This folder contains:</p> <ul> <li><strong>Data processing templates</strong> for normalization, aggregation, annotation, and feature selection.</li> <li>Templates for identifying cellular profiles using <strong>Linear Discriminant Analysis (LDA)</strong>.</li> <li><strong>Clustering methods</strong>, including Elbow Method, Silhouette Method, and K-means.</li> <li><strong>Confusion matrix analysis</strong> to evaluate clustering quality.</li> <li><strong>Feature importance identification</strong> scripts.</li> <li>Templates for <strong>statistical analysis</strong>, including Kruskal-Wallis tests with Bonferroni post hoc correction and boxplot visualization of individual features.</li> </ul> <hr> <p>These resources were developed to facilitate <strong>image-based workflows</strong> and advanced data analysis for understanding the hepatotoxic effects of microplastics in vitro.</p>