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

• Published in: bioRxiv
• Main Authors: Salek, Mahyar, Li, Nianzhen, Hou-Pu Chou, Saini, Kiran, Jovic, Andreja, Jacobs, Kevin B, Johnson, Chassidy, Lee, Esther J, Chang, Christina, Nguyen, Phuc, Mei, Jeanette, Pant, Krishna P, Wong-Thai, Amy Y, Smith, Quillan F, Huang, Stephanie, Chow, Ryan, Cruz, Janifer, Walker, Jeff, Chan, Bryan, Musci, Thomas J, Ashley, Euan A, Maddison (Mahdokht) Masaeli
• Format: Paper
• Language: English
• Published: Cold Spring Harbor Cold Spring Harbor Laboratory Press 07.03.2022
• Subjects: Cell morphology; Cytology; Deep learning; Gene expression; Morphology; Tumor cells
• DOI: 10.1101/2022.02.28.482368

Phenotyping of single cells has dramatically lagged advances in molecular characterization and remains a manual, subjective, and destructive process. We introduce COSMOS, a platform for phenotyping and enrichment of cells based on deep learning interpretation of high-content morphology data in realtime. By training models on an atlas of >1.5 billion images, we demonstrate enrichment of unlabeled cells up to 33,000 fold. We apply COSMOS to multicellular tissue biopsy samples demonstrating that it can identify malignant cells with similar accuracy to molecular approaches while enriching viable cells for functional evaluation. We show high-dimensional embedding vectors of morphology generated by COSMOS without any need for complex sample pre-processing, gating, or bioinformatics capabilities, which enables discovery of cellular phenotypes, and integration of morphology into multi-dimensional analyses. Competing Interest Statement All authors are current or former employees at or are affiliated with Deepcell Inc.