1
|
Carreras-Puigvert J, Spjuth O. Artificial intelligence for high content imaging in drug discovery. Curr Opin Struct Biol 2024; 87:102842. [PMID: 38797109 DOI: 10.1016/j.sbi.2024.102842] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/26/2024] [Revised: 04/28/2024] [Accepted: 04/29/2024] [Indexed: 05/29/2024]
Abstract
Artificial intelligence (AI) and high-content imaging (HCI) are contributing to advancements in drug discovery, propelled by the recent progress in deep neural networks. This review highlights AI's role in analysis of HCI data from fixed and live-cell imaging, enabling novel label-free and multi-channel fluorescent screening methods, and improving compound profiling. HCI experiments are rapid and cost-effective, facilitating large data set accumulation for AI model training. However, the success of AI in drug discovery also depends on high-quality data, reproducible experiments, and robust validation to ensure model performance. Despite challenges like the need for annotated compounds and managing vast image data, AI's potential in phenotypic screening and drug profiling is significant. Future improvements in AI, including increased interpretability and integration of multiple modalities, are expected to solidify AI and HCI's role in drug discovery.
Collapse
Affiliation(s)
- Jordi Carreras-Puigvert
- Department of Pharmaceutical Biosciences and Science for Life Laboratories, Uppsala University, Sweden.
| | - Ola Spjuth
- Department of Pharmaceutical Biosciences and Science for Life Laboratories, Uppsala University, Sweden.
| |
Collapse
|
2
|
Weitz P, Valkonen M, Solorzano L, Carr C, Kartasalo K, Boissin C, Koivukoski S, Kuusela A, Rasic D, Feng Y, Pouplier SS, Sharma A, Eriksson KL, Robertson S, Marzahl C, Gatenbee CD, Anderson ARA, Wodzinski M, Jurgas A, Marini N, Atzori M, Müller H, Budelmann D, Weiss N, Heldmann S, Lotz J, Wolterink JM, De Santi B, Patil A, Sethi A, Kondo S, Kasai S, Hirasawa K, Farrokh M, Kumar N, Greiner R, Latonen L, Laenkholm AV, Hartman J, Ruusuvuori P, Rantalainen M. The ACROBAT 2022 challenge: Automatic registration of breast cancer tissue. Med Image Anal 2024; 97:103257. [PMID: 38981282 DOI: 10.1016/j.media.2024.103257] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/05/2023] [Revised: 05/17/2024] [Accepted: 06/24/2024] [Indexed: 07/11/2024]
Abstract
The alignment of tissue between histopathological whole-slide-images (WSI) is crucial for research and clinical applications. Advances in computing, deep learning, and availability of large WSI datasets have revolutionised WSI analysis. Therefore, the current state-of-the-art in WSI registration is unclear. To address this, we conducted the ACROBAT challenge, based on the largest WSI registration dataset to date, including 4,212 WSIs from 1,152 breast cancer patients. The challenge objective was to align WSIs of tissue that was stained with routine diagnostic immunohistochemistry to its H&E-stained counterpart. We compare the performance of eight WSI registration algorithms, including an investigation of the impact of different WSI properties and clinical covariates. We find that conceptually distinct WSI registration methods can lead to highly accurate registration performances and identify covariates that impact performances across methods. These results provide a comparison of the performance of current WSI registration methods and guide researchers in selecting and developing methods.
Collapse
Affiliation(s)
- Philippe Weitz
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden.
| | - Masi Valkonen
- Institute of Biomedicine, University of Turku, Turku, Finland
| | - Leslie Solorzano
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden
| | - Circe Carr
- Institute of Biomedicine, University of Turku, Turku, Finland
| | - Kimmo Kartasalo
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden
| | - Constance Boissin
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden
| | - Sonja Koivukoski
- Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland
| | - Aino Kuusela
- Institute of Biomedicine, University of Turku, Turku, Finland
| | - Dusan Rasic
- Department of Surgical Pathology, Zealand University Hospital, Roskilde, Denmark
| | - Yanbo Feng
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden
| | | | - Abhinav Sharma
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden
| | - Kajsa Ledesma Eriksson
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden
| | - Stephanie Robertson
- Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden
| | | | - Chandler D Gatenbee
- Department of Integrated Mathematical Oncology, Moffitt Cancer Center, Tampa, USA
| | | | - Marek Wodzinski
- Informatics Institute, University of Applied Sciences Western Switzerland, Switzerland; Department of Measurement and Electronics, AGH University of Kraków, Poland
| | - Artur Jurgas
- Informatics Institute, University of Applied Sciences Western Switzerland, Switzerland; Department of Measurement and Electronics, AGH University of Kraków, Poland
| | - Niccolò Marini
- Informatics Institute, University of Applied Sciences Western Switzerland, Switzerland; Department of Computer Science, University of Geneva, Geneva, Switzerland
| | - Manfredo Atzori
- Informatics Institute, University of Applied Sciences Western Switzerland, Switzerland; Department of Neuroscience, University of Padova, Italy
| | - Henning Müller
- Informatics Institute, University of Applied Sciences Western Switzerland, Switzerland; Medical Faculty, University of Geneva, Switzerland
| | - Daniel Budelmann
- Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany
| | - Nick Weiss
- Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany
| | - Stefan Heldmann
- Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany
| | - Johannes Lotz
- Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany
| | - Jelmer M Wolterink
- Department of Applied Mathematics, Technical Medical Centre, University of Twente, Enschede, The Netherlands
| | - Bruno De Santi
- Multimodality Medical Imaging, Technical Medical Centre, University of Twente, Enschede, The Netherlands
| | - Abhijeet Patil
- Department of Electrical Engineering, Indian Institute of Technology, Bombay, India
| | - Amit Sethi
- Department of Electrical Engineering, Indian Institute of Technology, Bombay, India
| | - Satoshi Kondo
- Graduate School of Engineering, Muroran Institute of Technology, Hokkaido, Japan
| | - Satoshi Kasai
- Faculty of Medical Technology, Niigata University of Health and Welfare, Niigata, Japan
| | | | - Mahtab Farrokh
- Department of Computing Science, University of Alberta, Edmonton, Alberta
| | - Neeraj Kumar
- Department of Computing Science, University of Alberta, Edmonton, Alberta
| | - Russell Greiner
- Department of Computing Science, University of Alberta, Edmonton, Alberta; Alberta Machine Intelligence Institute, Edmonton, Canada
| | - Leena Latonen
- Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland
| | | | - Johan Hartman
- Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden; MedTechLabs, BioClinicum, Karolinska University Hospital, Stockholm, Sweden
| | - Pekka Ruusuvuori
- Institute of Biomedicine, University of Turku, Turku, Finland; Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland
| | - Mattias Rantalainen
- Department of Medical Epidemiology and Biostatistics, Karolinska Insitutet, Stockholm, Sweden; MedTechLabs, BioClinicum, Karolinska University Hospital, Stockholm, Sweden.
| |
Collapse
|
3
|
Grexa I, Iván ZZ, Migh E, Kovács F, Bolck HA, Zheng X, Mund A, Moshkov N, Miczán V, Koos K, Horvath P. SuperCUT, an unsupervised multimodal image registration with deep learning for biomedical microscopy. Brief Bioinform 2024; 25:bbae029. [PMID: 38483256 PMCID: PMC10938542 DOI: 10.1093/bib/bbae029] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/19/2023] [Revised: 12/20/2023] [Accepted: 01/08/2024] [Indexed: 03/17/2024] Open
Abstract
Numerous imaging techniques are available for observing and interrogating biological samples, and several of them can be used consecutively to enable correlative analysis of different image modalities with varying resolutions and the inclusion of structural or molecular information. Achieving accurate registration of multimodal images is essential for the correlative analysis process, but it remains a challenging computer vision task with no widely accepted solution. Moreover, supervised registration methods require annotated data produced by experts, which is limited. To address this challenge, we propose a general unsupervised pipeline for multimodal image registration using deep learning. We provide a comprehensive evaluation of the proposed pipeline versus the current state-of-the-art image registration and style transfer methods on four types of biological problems utilizing different microscopy modalities. We found that style transfer of modality domains paired with fully unsupervised training leads to comparable image registration accuracy to supervised methods and, most importantly, does not require human intervention.
Collapse
Affiliation(s)
- Istvan Grexa
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
- Doctoral School of Interdisciplinary Medicine, University of Szeged, Korányi fasor 10, Szeged 6720 Hungary
| | - Zsanett Zsófia Iván
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
- Doctoral School of Biology, University of Szeged, Közép fasor 52, Szeged 6726 Hungary
| | - Ede Migh
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
| | - Ferenc Kovács
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
- Single-Cell Technologies Ltd, Temesvári körút 62, Szeged 6726, Hungary
| | - Hella A Bolck
- Department of Pathology and Molecular Pathology, University Hospital Zürich, Zürich, Schmelzbergstrasse 12 8091, Switzerland
| | - Xiang Zheng
- Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Tuborg Havnevej 19 2900 Hellerup, Denmark
| | - Andreas Mund
- Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Tuborg Havnevej 19 2900 Hellerup, Denmark
| | - Nikita Moshkov
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
| | - Vivien Miczán
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
| | - Krisztian Koos
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
| | - Peter Horvath
- Synthetic and Systems Biology Unit, Biological Research Centre (BRC), Temesvári körút 62, Szeged 6726
- Single-Cell Technologies Ltd, Temesvári körút 62, Szeged 6726, Hungary
- Institute for Molecular Medicine Finland (FIMM), Helsinki Institute of Life Science, University of Helsinki, Tukholmankatu 8, Helsinki 00014, Finland
- Institute of AI for Health, Helmholtz Zentrum München, Ingolstädter Landstraße 1, 85764 Oberschleißheim Neuherberg, Germany
| |
Collapse
|
4
|
Timonen VA, Kerkelä E, Impola U, Penna L, Partanen J, Kilpivaara O, Arvas M, Pitkänen E. DeepIFC: Virtual fluorescent labeling of blood cells in imaging flow cytometry data with deep learning. Cytometry A 2023; 103:807-817. [PMID: 37276178 DOI: 10.1002/cyto.a.24770] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/30/2022] [Revised: 05/16/2023] [Accepted: 06/02/2023] [Indexed: 06/07/2023]
Abstract
Imaging flow cytometry (IFC) combines flow cytometry with microscopy, allowing rapid characterization of cellular and molecular properties via high-throughput single-cell fluorescent imaging. However, fluorescent labeling is costly and time-consuming. We present a computational method called DeepIFC based on the Inception U-Net neural network architecture, able to generate fluorescent marker images and learn morphological features from IFC brightfield and darkfield images. Furthermore, the DeepIFC workflow identifies cell types from the generated fluorescent images and visualizes the single-cell features generated in a 2D space. We demonstrate that rarer cell types are predicted well when a balanced data set is used to train the model, and the model is able to recognize red blood cells not seen during model training as a distinct entity. In summary, DeepIFC allows accurate cell reconstruction, typing and recognition of unseen cell types from brightfield and darkfield images via virtual fluorescent labeling.
Collapse
Affiliation(s)
- Veera A Timonen
- Institute for Molecular Medicine Finland (FIMM), Helsinki Institute of Life Science (HiLIFE), University of Helsinki, Helsinki, Finland
- Applied Tumor Genomics Research Program, Research Programs Unit, Faculty of Medicine, University of Helsinki, Helsinki, Finland
| | - Erja Kerkelä
- Advanced Cell Therapy Centre, Finnish Red Cross Blood Service, Vantaa, Finland
| | - Ulla Impola
- Research and Development, Finnish Red Cross Blood Service, Helsinki, Finland
| | - Leena Penna
- Research and Development, Finnish Red Cross Blood Service, Helsinki, Finland
| | - Jukka Partanen
- Research and Development, Finnish Red Cross Blood Service, Helsinki, Finland
| | - Outi Kilpivaara
- Applied Tumor Genomics Research Program, Research Programs Unit, Faculty of Medicine, University of Helsinki, Helsinki, Finland
- Department of Medical and Clinical Genetics, Medicum, Faculty of Medicine, University of Helsinki, Helsinki, Finland
- HUSLAB Laboratory of Genetics, HUS Diagnostic Center, Helsinki University Hospital, Helsinki, Finland
- iCAN Digital Precision Cancer Medicine Flagship, Helsinki, Finland
| | - Mikko Arvas
- Research and Development, Finnish Red Cross Blood Service, Helsinki, Finland
| | - Esa Pitkänen
- Institute for Molecular Medicine Finland (FIMM), Helsinki Institute of Life Science (HiLIFE), University of Helsinki, Helsinki, Finland
- Applied Tumor Genomics Research Program, Research Programs Unit, Faculty of Medicine, University of Helsinki, Helsinki, Finland
- iCAN Digital Precision Cancer Medicine Flagship, Helsinki, Finland
| |
Collapse
|
5
|
Weitz P, Valkonen M, Solorzano L, Carr C, Kartasalo K, Boissin C, Koivukoski S, Kuusela A, Rasic D, Feng Y, Sinius Pouplier S, Sharma A, Ledesma Eriksson K, Latonen L, Laenkholm AV, Hartman J, Ruusuvuori P, Rantalainen M. A Multi-Stain Breast Cancer Histological Whole-Slide-Image Data Set from Routine Diagnostics. Sci Data 2023; 10:562. [PMID: 37620357 PMCID: PMC10449765 DOI: 10.1038/s41597-023-02422-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/16/2023] [Accepted: 07/27/2023] [Indexed: 08/26/2023] Open
Abstract
The analysis of FFPE tissue sections stained with haematoxylin and eosin (H&E) or immunohistochemistry (IHC) is essential for the pathologic assessment of surgically resected breast cancer specimens. IHC staining has been broadly adopted into diagnostic guidelines and routine workflows to assess the status of several established biomarkers, including ER, PGR, HER2 and KI67. Biomarker assessment can also be facilitated by computational pathology image analysis methods, which have made numerous substantial advances recently, often based on publicly available whole slide image (WSI) data sets. However, the field is still considerably limited by the sparsity of public data sets. In particular, there are no large, high quality publicly available data sets with WSIs of matching IHC and H&E-stained tissue sections from the same tumour. Here, we publish the currently largest publicly available data set of WSIs of tissue sections from surgical resection specimens from female primary breast cancer patients with matched WSIs of corresponding H&E and IHC-stained tissue, consisting of 4,212 WSIs from 1,153 patients.
Collapse
Affiliation(s)
- Philippe Weitz
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
| | - Masi Valkonen
- Institute of Biomedicine, University of Turku, Turku, Finland
| | - Leslie Solorzano
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
| | - Circe Carr
- Institute of Biomedicine, University of Turku, Turku, Finland
| | - Kimmo Kartasalo
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
| | - Constance Boissin
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
| | - Sonja Koivukoski
- Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland
| | - Aino Kuusela
- Institute of Biomedicine, University of Turku, Turku, Finland
| | - Dusan Rasic
- Department of Surgical Pathology, Zealand University Hospital, Roskilde, Denmark
- Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
| | - Yanbo Feng
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
| | - Sandra Sinius Pouplier
- Department of Surgical Pathology, Zealand University Hospital, Roskilde, Denmark
- Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
| | - Abhinav Sharma
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
| | - Kajsa Ledesma Eriksson
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
| | - Leena Latonen
- Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland
- Foundation for the Finnish Cancer Institute, Helsinki, Finland
| | - Anne-Vibeke Laenkholm
- Department of Surgical Pathology, Zealand University Hospital, Roskilde, Denmark
- Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
| | - Johan Hartman
- Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden
- MedTechLabs, BioClinicum, Karolinska University Hospital, Stockholm, Sweden
| | - Pekka Ruusuvuori
- Institute of Biomedicine, University of Turku, Turku, Finland
- Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland
| | - Mattias Rantalainen
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
- MedTechLabs, BioClinicum, Karolinska University Hospital, Stockholm, Sweden.
| |
Collapse
|
6
|
Harrison PJ, Gupta A, Rietdijk J, Wieslander H, Carreras-Puigvert J, Georgiev P, Wählby C, Spjuth O, Sintorn IM. Evaluating the utility of brightfield image data for mechanism of action prediction. PLoS Comput Biol 2023; 19:e1011323. [PMID: 37490493 PMCID: PMC10403126 DOI: 10.1371/journal.pcbi.1011323] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/27/2022] [Revised: 08/04/2023] [Accepted: 07/02/2023] [Indexed: 07/27/2023] Open
Abstract
Fluorescence staining techniques, such as Cell Painting, together with fluorescence microscopy have proven invaluable for visualizing and quantifying the effects that drugs and other perturbations have on cultured cells. However, fluorescence microscopy is expensive, time-consuming, labor-intensive, and the stains applied can be cytotoxic, interfering with the activity under study. The simplest form of microscopy, brightfield microscopy, lacks these downsides, but the images produced have low contrast and the cellular compartments are difficult to discern. Nevertheless, by harnessing deep learning, these brightfield images may still be sufficient for various predictive purposes. In this study, we compared the predictive performance of models trained on fluorescence images to those trained on brightfield images for predicting the mechanism of action (MoA) of different drugs. We also extracted CellProfiler features from the fluorescence images and used them to benchmark the performance. Overall, we found comparable and largely correlated predictive performance for the two imaging modalities. This is promising for future studies of MoAs in time-lapse experiments for which using fluorescence images is problematic. Explorations based on explainable AI techniques also provided valuable insights regarding compounds that were better predicted by one modality over the other.
Collapse
Affiliation(s)
- Philip John Harrison
- Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
- Science for Life Laboratory, Uppsala, Sweden
| | - Ankit Gupta
- Department of Information Technology, Uppsala University, Uppsala, Sweden
| | - Jonne Rietdijk
- Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
- Science for Life Laboratory, Uppsala, Sweden
| | - Håkan Wieslander
- Department of Information Technology, Uppsala University, Uppsala, Sweden
| | - Jordi Carreras-Puigvert
- Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
- Science for Life Laboratory, Uppsala, Sweden
| | - Polina Georgiev
- Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
- Science for Life Laboratory, Uppsala, Sweden
| | - Carolina Wählby
- Science for Life Laboratory, Uppsala, Sweden
- Department of Information Technology, Uppsala University, Uppsala, Sweden
| | - Ola Spjuth
- Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
- Science for Life Laboratory, Uppsala, Sweden
| | - Ida-Maria Sintorn
- Department of Information Technology, Uppsala University, Uppsala, Sweden
| |
Collapse
|
7
|
Label-free prediction of cell painting from brightfield images. Sci Rep 2022; 12:10001. [PMID: 35705591 PMCID: PMC9200748 DOI: 10.1038/s41598-022-12914-x] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/05/2021] [Accepted: 05/18/2022] [Indexed: 11/08/2022] Open
Abstract
Cell Painting is a high-content image-based assay applied in drug discovery to predict bioactivity, assess toxicity and understand mechanisms of action of chemical and genetic perturbations. We investigate label-free Cell Painting by predicting the five fluorescent Cell Painting channels from brightfield input. We train and validate two deep learning models with a dataset representing 17 batches, and we evaluate on batches treated with compounds from a phenotypic set. The mean Pearson correlation coefficient of the predicted images across all channels is 0.84. Without incorporating features into the model training, we achieved a mean correlation of 0.45 with ground truth features extracted using a segmentation-based feature extraction pipeline. Additionally, we identified 30 features which correlated greater than 0.8 to the ground truth. Toxicity analysis on the label-free Cell Painting resulted a sensitivity of 62.5% and specificity of 99.3% on images from unseen batches. We provide a breakdown of the feature profiles by channel and feature type to understand the potential and limitations of label-free morphological profiling. We demonstrate that label-free Cell Painting has the potential to be used for downstream analyses and could allow for repurposing imaging channels for other non-generic fluorescent stains of more targeted biological interest.
Collapse
|