HCS Methodology for Helping in Lab Scale Image-Based Assays

Abstract

High-content screening (HCS) automates image acquisition and analysis in microscopy. This technology considers the multiple parameters contained in the images and produces statistically significant results. The recent improvements in image acquisition throughput, image analysis, and machine learning (ML) have popularized this kind of experiments, emphasizing the need for new tools and know-how to help in its design, analysis, and data interpretation. This chapter summarizes HCS recommendations for lab scale assays and provides both macros for HCS-oriented image analysis and user-friendly tools for data mining processes. All the steps described herein are oriented to a wide variety of image cell-based experiments. The workflows are illustrated with practical examples and test images. Their use is expected to help analyze thousands of images, create graphical representations, and apply machine learning models on HCS.

Keywords: Automated analysis; Data mining; HCS; ImageJ macro language; Machine learning.