Self-supervised deep learning uncovers the semantic landscape of drug-induced latent mitochondrial phenotypes
Preprint
Abstract
SUMMARY: Imaging-based high-content screening aims to identify substances that modulate cellular phenotypes. Traditional approaches screen compounds for their ability to shift disease phenotypes toward healthy phenotypes, but these end point-based screens lack an atlas-like mapping between phenotype and cell state that covers the full spectrum of possible phenotypic responses. In this study, we present MitoSpace: a novel mitochondrial phenotypic atlas that leverages self-supervised deep learning to create a semantically meaningful latent space from images without relying on any data labels for training. Our approach employs a dataset of ∼100,000 microscopy images of Cal27 and HeLa cells treated with 25 drugs affecting mitochondria, but can be generalized to any cell type, cell organelle, or drug library with no changes to the methodology. We demonstrate how MitoSpace enhances our understanding of the range of mitochondrial phenotypes induced by pharmacological interventions. We find that i) self-supervised learning can automatically uncover the semantic landscape of drug induced latent mitochondrial phenotypes and can map individual cells to the correct functional area of the drug they are treated with, ii) the traditional classification of mitochondrial morphology along a fragmented to fused axis is more complex than previously thought, with additional axes being identified, and iii) latent spaces trained in a self-supervised manner are superior to those trained with supervised models, and generalize to other cell types and drug conditions without explicit training on those cell types or drug conditions. Future applications of MitoSpace include creating mitochondrial biomarkers for drug discovery and determining the effects of unknown drugs and diseases for diagnostic purposes.Keywords
This publication has 62 references indexed in Scilit:
- Use of artificial intelligence to enhance phenotypic drug discoveryDrug Discovery Today, 2021
- Pre-trained models for natural language processing: A surveyScience China Technological Sciences, 2020
- A High-Content Screen Identifies TPP1 and Aurora B as Regulators of Axonal Mitochondrial TransportCell Reports, 2019
- Mitochondria as pharmacological targets in Down syndromeFree Radical Biology and Medicine, 2018
- Mitochondria and ageing: role in heart, skeletal muscle and adipose tissueJournal of Cachexia, Sarcopenia and Muscle, 2017
- A Hierarchical Map of Regulatory Genetic Interactions in Membrane TraffickingCell, 2014
- Integration of Chemical and RNAi Multiparametric Profiles Identifies Triggers of Intracellular Mycobacterial KillingCell Host & Microbe, 2013
- Morphobase, an Encyclopedic Cell Morphology Database, and Its Use for Drug Target IdentificationChemistry & Biology, 2012
- The beautiful cell: high-content screening in drug discoveryAnalytical and Bioanalytical Chemistry, 2010
- Two new human tumor cell lines derived from squamous cell carcinomas of the tongue: Establishment, characterization and response to cytotoxic treatmentEuropean Journal of Cancer and Clinical Oncology, 1988