High-throughput deconvolution of 3D organoid dynamics at cellular resolution for cancer pharmacology with Cellos

TLDR

Three‑dimensional organoid cultures enable interrogation of cellular growth, morphology, spatial architecture, and interactions in response to treatment. We aim to develop a computational method that delivers high‑throughput, cellular‑resolution analysis of 3D organoids. Cellos combines classical segmentation algorithms with a Stardist‑3D convolutional neural network to accurately segment nuclei and distinguish labeled cell populations across ~100,000 organoids and 2.35 million cells. Cellos reproduces luminescence‑based drug responses while quantifying organoid and nuclear morphology changes and cell‑cell spatial relationships, offering a powerful tool for high‑throughput pharmacological testing and biological investigation.

Abstract

Abstract Three-dimensional (3D) organoid cultures are flexible systems to interrogate cellular growth, morphology, multicellular spatial architecture, and cellular interactions in response to treatment. However, computational methods for analysis of 3D organoids with sufficiently high-throughput and cellular resolution are needed. Here we report Cellos, an accurate, high-throughput pipeline for 3D organoid segmentation using classical algorithms and nuclear segmentation using a trained Stardist-3D convolutional neural network. To evaluate Cellos, we analyze ~100,000 organoids with ~2.35 million cells from multiple treatment experiments. Cellos segments dye-stained or fluorescently-labeled nuclei and accurately distinguishes distinct labeled cell populations within organoids. Cellos can recapitulate traditional luminescence-based drug response of cells with complex drug sensitivities, while also quantifying changes in organoid and nuclear morphologies caused by treatment as well as cell-cell spatial relationships that reflect ecological affinity. Cellos provides powerful tools to perform high-throughput analysis for pharmacological testing and biological investigation of organoids based on 3D imaging.