High-speed fluorescence image–enabled cell sorting

Abstract

Fast and selective isolation of single cells with unique spatial and morphological traits remains a technical challenge. Here, we address this by establishing high-speed image-enabled cell sorting (ICS), which records multicolor fluorescence images and sorts cells based on measurements from image data at speeds up to 15,000 events per second. We show that ICS quantifies cell morphology and localization of labeled proteins and increases the resolution of cell cycle analyses by separating mitotic stages. We combine ICS with CRISPR-pooled screens to identify regulators of the nuclear factor κB (NF-κB) pathway, enabling the completion of genome-wide image-based screens in about 9 hours of run time. By assessing complex cellular phenotypes, ICS substantially expands the phenotypic space accessible to cell-sorting applications and pooled genetic screening.

Realtime morphological characterization and sorting of unlabeled viable cells using deep learning

“…Additionally, these approaches alter cells making them non-ideal for downstream characterization (Bendall et al 2011; Bendall et al 2012). There have been recent efforts to improve upon our capability to isolate cells based on their morphological traits (Schraivogel et al 2022), but these approaches still rely on staining cells with fluorescent markers, which alters them. Additionally, they are limited by the number of morphological traits that can be visualized simultaneously and require heavily involved processes to define a small number of features to quantify morphology.…”

“…In conclusion, we present COSMOS, a novel technology platform for the characterization, classification, isolation, and enrichment of cells from living organisms based on high-dimensional morphology. Recent work has motivated morphology as an analyte in cell sorting (Schraivogel et al 2022). Here we capture the power of deep neural networks in processing morphology by amassing an annotated atlas of greater than 1.5 billion single cell images and training deep models with the computational capacity to classify high resolution high content images.…”

Sorting single T cells based on secreted cytokines and surface markers using hydrogel nanovials

“…Therefore, the area/height ratio, or other metrics of fluorescence peak width or shape, can be used to distinguish these different states, unlike other approaches like ICS or the Miltenyi cytokine catch assay in which all staining occurs on the cell surface or intracellularly and cannot be distinguished 13,[25][26][27][28] . Images from image cytometry systems 29,30 or imageactivated cell sorting systems 31,32 would provide the similar ability to distinguish between cell surface staining and staining on nanovials.…”

The Next Generation of FACS – Not Just a Uniform Blob of Fluorescence

“…1 Fast forward to today, and we have dozens of FACS machines that can separate tens of thousands of cells per second based on combinatorial expression of >15 markers and some that even press beyond this. But scale is not the only thing that matters and a recent article by Schraivogel et al 2 highlights a new FACS-based technology that launches us into the next galaxy of possibilities-spatial sorting.…”

“…In his original article about the cell sorter published in 1965, Mack Fulwyler imagined “it may be possible to also measure simultaneously two (or more) characteristics of a cell and to make separation dependent on the ratio of such characteristics.” 1 Fast forward to today, and we have dozens of FACS machines that can separate tens of thousands of cells per second based on combinatorial expression of >15 markers and some that even press beyond this. But scale is not the only thing that matters and a recent article by Schraivogel et al 2 highlights a new FACS-based technology that launches us into the next galaxy of possibilities—spatial sorting.…”