In image-based profiling, software extracts thousands of morphological features of cells from multi-channel fluorescence microscopy images, yielding single-cell profiles that can be used for basic research and drug discovery. Powerful applications have been proven, including clustering chemical and genetic perturbations based on their similar morphological impact, identifying disease phenotypes by observing differences in profiles between healthy and diseased cells, and predicting assay outcomes using machine learning, among many others. Here we provide an updated protocol for the most popular assay for image-based profiling, Cell Painting. Introduced in 2013, it uses six stains imaged in five channels and labels eight diverse components of the cell: DNA, cytoplasmic RNA, nucleoli, actin, golgi apparatus, plasma membrane, endoplasmic reticulum, and mitochondria. The original protocol was updated in 2016 based on several years' experience running it at two sites, after optimizing it by visual stain quality. Here we describe the work of the Joint Undertaking for Morphological Profiling (JUMP) Cell Painting Consortium, aiming to improve upon the assay via quantitative optimization, based on the measured ability of the assay to detect morphological phenotypes and group similar perturbations together. We find that the assay gives very robust outputs despite a variety of changes to the protocol and that two vendors' dyes work equivalently well. We present Cell Painting version 3, in which some steps are simplified and several stain concentrations can be reduced, saving costs. Cell culture and image acquisition take 1 to 2 weeks for a typically sized batch of 20 or fewer plates; feature extraction and data analysis take an additional 1 to 2 weeks.

We present a new, carefully designed and well-annotated dataset of images and image-based profiles of cells that have been treated with chemical compounds and genetic perturbations. Each gene that is perturbed is a known target of at least two compounds in the dataset. The dataset can thus serve as a benchmark to evaluate methods for predicting similarities between compounds and between genes and compounds, measuring the effect size of a perturbation, and more generally, learning effective representations for measuring cellular state from microscopy images. Advancements in these applications can accelerate the development of new medicines.

In image-based profiling, software extracts thousands of morphological features of cells from multi-channel fluorescence microscopy images, yielding single-cell profiles that can be used for basic research and drug discovery. Powerful applications have been proven, including clustering chemical and genetic perturbations based on their similar morphological impact, identifying disease phenotypes by observing differences in profiles between healthy and diseased cells, and predicting assay outcomes using machine learning, among many others. Here we provide an updated protocol for the most popular assay for image-based profiling, Cell Painting. Introduced in 2013, it uses six stains imaged in five channels and labels eight diverse components of the cell: DNA, cytoplasmic RNA, nucleoli, actin, golgi apparatus, plasma membrane, endoplasmic reticulum, and mitochondria. The original protocol was updated in 2016 based on several years' experience running it at two sites, after optimizing it by visual stain quality. Here we describe the work of the Joint Undertaking for Morphological Profiling (JUMP) Cell Painting Consortium, aiming to improve upon the assay via quantitative optimization, based on the measured ability of the assay to detect morphological phenotypes and group similar perturbations together. We find that the assay gives very robust outputs despite a variety of changes to the protocol and that two vendors' dyes work equivalently well. We present Cell Painting version 3, in which some steps are simplified and several stain concentrations can be reduced, saving costs. Cell culture and image acquisition take 1-2 weeks for a typically sized batch of 20 or fewer plates; feature extraction and data analysis take an additional 1-2 weeks..

Single-cell resolution technologies warrant computational methods that capture cell heterogeneity while allowing efficient comparisons of populations. Here, we summarize cell populations by adding features’ dispersion and covariances to population averages, in the context of image-based profiling. We find that data fusion is critical for these metrics to improve results over the prior alternatives, providing at least ~20% better performance in predicting a compound’s mechanism of action (MoA) and a gene’s pathway.

Cellular exposure to free fatty acids (FFA) is implicated in the pathogenesis of obesity-associated diseases. However, studies to date have assumed that a few select FFAs are representative of broad structural categories, and there are no scalable approaches to comprehensively assess the biological processes induced by exposure to diverse FFAs circulating in human plasma. Furthermore, assessing how these FFA-mediated processes interact with genetic risk for disease remains elusive. Here we report the design and implementation of FALCON (Fatty Acid Library for Comprehensive ONtologies) as an unbiased, scalable and multimodal interrogation of 61 structurally diverse FFAs. We identified a subset of lipotoxic monounsaturated fatty acids (MUFAs) with a distinct lipidomic profile associated with decreased membrane fluidity. Furthermore, we developed a new approach to prioritize genes that reflect the combined effects of exposure to harmful FFAs and genetic risk for type 2 diabetes (T2D). Importantly, we found that c-MAF inducing protein (CMIP) protects cells from exposure to FFAs by modulating Akt signaling and we validated the role of CMIP in human pancreatic beta cells. In sum, FALCON empowers the study of fundamental FFA biology and offers an integrative approach to identify much needed targets for diverse diseases associated with disordered FFA metabolism.