22
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Learning unsupervised feature representations for single cell microscopy images with paired cell inpainting

      research-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Cellular microscopy images contain rich insights about biology. To extract this information, researchers use features, or measurements of the patterns of interest in the images. Here, we introduce a convolutional neural network (CNN) to automatically design features for fluorescence microscopy. We use a self-supervised method to learn feature representations of single cells in microscopy images without labelled training data. We train CNNs on a simple task that leverages the inherent structure of microscopy images and controls for variation in cell morphology and imaging: given one cell from an image, the CNN is asked to predict the fluorescence pattern in a second different cell from the same image. We show that our method learns high-quality features that describe protein expression patterns in single cells both yeast and human microscopy datasets. Moreover, we demonstrate that our features are useful for exploratory biological analysis, by capturing high-resolution cellular components in a proteome-wide cluster analysis of human proteins, and by quantifying multi-localized proteins and single-cell variability. We believe paired cell inpainting is a generalizable method to obtain feature representations of single cells in multichannel microscopy images.

          Author summary

          To understand the cell biology captured by microscopy images, researchers use features, or measurements of relevant properties of cells, such as the shape or size of cells, or the intensity of fluorescent markers. Features are the starting point of most image analysis pipelines, so their quality in representing cells is fundamental to the success of an analysis. Classically, researchers have relied on features manually defined by imaging experts. In contrast, deep learning techniques based on convolutional neural networks (CNNs) automatically learn features, which can outperform manually-defined features at image analysis tasks. However, most CNN methods require large manually-annotated training datasets to learn useful features, limiting their practical application. Here, we developed a new CNN method that learns high-quality features for single cells in microscopy images, without the need for any labeled training data. We show that our features surpass other comparable features in identifying protein localization from images, and that our method can generalize to diverse datasets. By exploiting our method, researchers will be able to automatically obtain high-quality features customized to their own image datasets, facilitating many downstream analyses, as we highlight by demonstrating many possible use cases of our features in this study.

          Related collections

          Most cited references33

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          Data-analysis strategies for image-based cell profiling

          This Review covers the steps required to create high-quality image-based profiles from high-throughput microscopy images.
            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            A brief introduction to weakly supervised learning

              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Dissecting DNA damage response pathways by analyzing protein localization and abundance changes during DNA replication stress

              Re-localization of proteins is a hallmark of the DNA damage response. We use high-throughput microscopic screening of the yeast GFP fusion collection to develop a systems-level view of protein re-organization following drug-induced DNA replication stress. Changes in protein localization and abundance reveal drug-specific patterns of functional enrichments. Classification of proteins by sub-cellular destination allows the identification of pathways that respond to replication stress. We analyzed pairwise combinations of GFP fusions and gene deletion mutants to define and order two novel DNA damage responses. In the first, Cmr1 forms subnuclear foci that are regulated by the histone deacetylase Hos2 and are distinct from the typical Rad52 repair foci. In a second example, we find that the checkpoint kinases Mec1/Tel1 and the translation regulator Asc1 regulate P-body formation. This method identifies response pathways that were not detected in genetic and protein interaction screens, and can be readily applied to any form of chemical or genetic stress to reveal cellular response pathways.
                Bookmark

                Author and article information

                Contributors
                Role: ConceptualizationRole: InvestigationRole: MethodologyRole: SoftwareRole: Writing – original draftRole: Writing – review & editing
                Role: Supervision
                Role: Supervision
                Role: ConceptualizationRole: Project administrationRole: ResourcesRole: SupervisionRole: Writing – review & editing
                Role: Editor
                Journal
                PLoS Comput Biol
                PLoS Comput. Biol
                plos
                ploscomp
                PLoS Computational Biology
                Public Library of Science (San Francisco, CA USA )
                1553-734X
                1553-7358
                3 September 2019
                September 2019
                : 15
                : 9
                : e1007348
                Affiliations
                [1 ] Department of Computer Science, University of Toronto, Toronto, Canada
                [2 ] Phenomic AI, Toronto, Canada
                [3 ] Department of Cell and Systems Biology, University of Toronto, Toronto, Canada
                [4 ] Center for Analysis of Genome Evolution and Function, University of Toronto, Toronto, Canada
                UNITED STATES
                Author notes

                Oren Z Kraus and Sam Cooper are employees of Phenomic AI.

                Author information
                http://orcid.org/0000-0002-6328-9492
                http://orcid.org/0000-0003-3118-3121
                Article
                PCOMPBIOL-D-19-00084
                10.1371/journal.pcbi.1007348
                6743779
                31479439
                1a389302-37a9-4d42-80a1-543382101110
                © 2019 Lu et al

                This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

                History
                : 15 January 2019
                : 20 August 2019
                Page count
                Figures: 6, Tables: 3, Pages: 27
                Funding
                This work was conducted on a GPU generously provided by Nvidia through their academic seeding grant. This work was funded by the National Science and Engineering Research Council (Pre-Doctoral Award), Canada Research Chairs (Tier II Chair), and the Canadian Foundation for Innovation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
                Categories
                Research Article
                Biology and Life Sciences
                Agriculture
                Crop Science
                Crops
                Research and Analysis Methods
                Imaging Techniques
                Fluorescence Imaging
                Engineering and Technology
                Signal Processing
                Image Processing
                Research and Analysis Methods
                Imaging Techniques
                Biology and Life Sciences
                Neuroscience
                Cognitive Science
                Cognitive Psychology
                Learning
                Biology and Life Sciences
                Psychology
                Cognitive Psychology
                Learning
                Social Sciences
                Psychology
                Cognitive Psychology
                Learning
                Biology and Life Sciences
                Neuroscience
                Learning and Memory
                Learning
                Research and Analysis Methods
                Microscopy
                Light Microscopy
                Fluorescence Microscopy
                Biology and Life Sciences
                Biochemistry
                Proteins
                Luminescent Proteins
                Green Fluorescent Protein
                Research and Analysis Methods
                Animal Studies
                Experimental Organism Systems
                Yeast and Fungal Models
                Custom metadata
                vor-update-to-uncorrected-proof
                2019-09-13
                The cyclops data is downloadable from http://cyclops.ccbr.utoronto.ca. The HPA data is downloadable from https://www.proteinatlas.org/humanproteome/cell, and a script for batch download can be found at https://github.com/alexxijielu/paired_cell_inpainting/blob/master/human_model/data_download/download_hpa.py. Single cell features and additional data are downloadable at http://hershey.csb.utoronto.ca/paired_cell_inpainting_features/.

                Quantitative & Systems biology
                Quantitative & Systems biology

                Comments

                Comment on this article