No Cover Image

Journal article 139 views 25 downloads

OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration

Bethany Hunter, Ioana Nicorescu, Emma Foster, David McDonald, Gillian Hulme, Andrew Fuller, Amanda Thomson, Thibaut Goldsborough, Catharien M. U. Hilkens, Joaquim Majo, Luke Milross, Andrew Fisher, Peter Bankhead, John Wills Orcid Logo, Paul Rees Orcid Logo, Andrew Filby, George Merces

Cytometry Part A, Volume: 105, Issue: 1, Pages: 36 - 53

Swansea University Author: Paul Rees Orcid Logo

  • 68202.VoR.pdf

    PDF | Version of Record

    © 2023 The Authors. This is an open access article under the terms of the Creative Commons Attribution License.

    Download (4.97MB)

Check full text

DOI (Published version): 10.1002/cyto.a.24803

Abstract

Analysis of imaging mass cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single-cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states,...

Full description

Published in: Cytometry Part A
ISSN: 1552-4922 1552-4930
Published: Wiley 2024
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa68202
first_indexed 2024-11-25T14:21:38Z
last_indexed 2025-01-16T20:49:09Z
id cronfa68202
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-01-16T12:27:28.7374139</datestamp><bib-version>v2</bib-version><id>68202</id><entry>2024-11-06</entry><title>OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration</title><swanseaauthors><author><sid>537a2fe031a796a3bde99679ee8c24f5</sid><ORCID>0000-0002-7715-6914</ORCID><firstname>Paul</firstname><surname>Rees</surname><name>Paul Rees</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-11-06</date><deptcode>EAAS</deptcode><abstract>Analysis of imaging mass cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single-cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states, or spatial relationships compared to reference data from single-cell suspension technologies. To this end we have developed the &#x201C;OPTimized Imaging Mass cytometry AnaLysis (OPTIMAL)&#x201D; framework to benchmark any approaches for cell segmentation, parameter transformation, batch effect correction, data visualization/clustering, and spatial neighborhood analysis. Using a panel of 27 metal-tagged antibodies recognizing well-characterized phenotypic and functional markers to stain the same Formalin-Fixed Paraffin Embedded (FFPE) human tonsil sample tissue microarray over 12 temporally distinct batches we tested several cell segmentation models, a range of different arcsinh cofactor parameter transformation values, 5 different dimensionality reduction algorithms, and 2 clustering methods. Finally, we assessed the optimal approach for performing neighborhood analysis. We found that single-cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using &#x201C;classical&#x201D; bivariate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximized the statistical separation between negative and positive signal distributions and a simple Z-score normalization step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing phenograph in terms of cell type identification. We also found that neighborhood analysis was influenced by the method used for finding neighboring cells with a &#x201C;disc&#x201D; pixel expansion outperforming a &#x201C;bounding box&#x201D; approach combined with the need for filtering objects based on size and image-edge location. Importantly, OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output and allows for single-cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists.</abstract><type>Journal Article</type><journal>Cytometry Part A</journal><volume>105</volume><journalNumber>1</journalNumber><paginationStart>36</paginationStart><paginationEnd>53</paginationEnd><publisher>Wiley</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1552-4922</issnPrint><issnElectronic>1552-4930</issnElectronic><keywords>image analysis; image cytometry; imaging mass cytometry; tissue segmentation</keywords><publishedDay>1</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-01-01</publishedDate><doi>10.1002/cyto.a.24803</doi><url/><notes/><college>COLLEGE NANME</college><department>Engineering and Applied Sciences School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>EAAS</DepartmentCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><funders>Medical Research Council UK Research and Innovations / NIHR UK Coronavirus Immunology Consortium. Grant Number: MR/V028448 European Union's Horizon 2020 research and innovation program. Grant Number: 860003 JGW Patterson Foundation United Kingdom Research and Innovation. Grant Number: EP/S02431X/1</funders><projectreference/><lastEdited>2025-01-16T12:27:28.7374139</lastEdited><Created>2024-11-06T19:32:22.7169219</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Engineering and Applied Sciences - Biomedical Engineering</level></path><authors><author><firstname>Bethany</firstname><surname>Hunter</surname><order>1</order></author><author><firstname>Ioana</firstname><surname>Nicorescu</surname><order>2</order></author><author><firstname>Emma</firstname><surname>Foster</surname><order>3</order></author><author><firstname>David</firstname><surname>McDonald</surname><order>4</order></author><author><firstname>Gillian</firstname><surname>Hulme</surname><order>5</order></author><author><firstname>Andrew</firstname><surname>Fuller</surname><order>6</order></author><author><firstname>Amanda</firstname><surname>Thomson</surname><order>7</order></author><author><firstname>Thibaut</firstname><surname>Goldsborough</surname><order>8</order></author><author><firstname>Catharien M. U.</firstname><surname>Hilkens</surname><order>9</order></author><author><firstname>Joaquim</firstname><surname>Majo</surname><order>10</order></author><author><firstname>Luke</firstname><surname>Milross</surname><order>11</order></author><author><firstname>Andrew</firstname><surname>Fisher</surname><order>12</order></author><author><firstname>Peter</firstname><surname>Bankhead</surname><order>13</order></author><author><firstname>John</firstname><surname>Wills</surname><orcid>0000-0002-4347-5394</orcid><order>14</order></author><author><firstname>Paul</firstname><surname>Rees</surname><orcid>0000-0002-7715-6914</orcid><order>15</order></author><author><firstname>Andrew</firstname><surname>Filby</surname><order>16</order></author><author><firstname>George</firstname><surname>Merces</surname><order>17</order></author></authors><documents><document><filename>68202__33356__74e079d169674d4c88c8e7130f5286d2.pdf</filename><originalFilename>68202.VoR.pdf</originalFilename><uploaded>2025-01-16T12:25:34.6264025</uploaded><type>Output</type><contentLength>5214989</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2023 The Authors. This is an open access article under the terms of the Creative Commons Attribution License.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2025-01-16T12:27:28.7374139 v2 68202 2024-11-06 OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration 537a2fe031a796a3bde99679ee8c24f5 0000-0002-7715-6914 Paul Rees Paul Rees true false 2024-11-06 EAAS Analysis of imaging mass cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single-cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states, or spatial relationships compared to reference data from single-cell suspension technologies. To this end we have developed the “OPTimized Imaging Mass cytometry AnaLysis (OPTIMAL)” framework to benchmark any approaches for cell segmentation, parameter transformation, batch effect correction, data visualization/clustering, and spatial neighborhood analysis. Using a panel of 27 metal-tagged antibodies recognizing well-characterized phenotypic and functional markers to stain the same Formalin-Fixed Paraffin Embedded (FFPE) human tonsil sample tissue microarray over 12 temporally distinct batches we tested several cell segmentation models, a range of different arcsinh cofactor parameter transformation values, 5 different dimensionality reduction algorithms, and 2 clustering methods. Finally, we assessed the optimal approach for performing neighborhood analysis. We found that single-cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using “classical” bivariate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximized the statistical separation between negative and positive signal distributions and a simple Z-score normalization step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing phenograph in terms of cell type identification. We also found that neighborhood analysis was influenced by the method used for finding neighboring cells with a “disc” pixel expansion outperforming a “bounding box” approach combined with the need for filtering objects based on size and image-edge location. Importantly, OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output and allows for single-cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists. Journal Article Cytometry Part A 105 1 36 53 Wiley 1552-4922 1552-4930 image analysis; image cytometry; imaging mass cytometry; tissue segmentation 1 1 2024 2024-01-01 10.1002/cyto.a.24803 COLLEGE NANME Engineering and Applied Sciences School COLLEGE CODE EAAS Swansea University Another institution paid the OA fee Medical Research Council UK Research and Innovations / NIHR UK Coronavirus Immunology Consortium. Grant Number: MR/V028448 European Union's Horizon 2020 research and innovation program. Grant Number: 860003 JGW Patterson Foundation United Kingdom Research and Innovation. Grant Number: EP/S02431X/1 2025-01-16T12:27:28.7374139 2024-11-06T19:32:22.7169219 Faculty of Science and Engineering School of Engineering and Applied Sciences - Biomedical Engineering Bethany Hunter 1 Ioana Nicorescu 2 Emma Foster 3 David McDonald 4 Gillian Hulme 5 Andrew Fuller 6 Amanda Thomson 7 Thibaut Goldsborough 8 Catharien M. U. Hilkens 9 Joaquim Majo 10 Luke Milross 11 Andrew Fisher 12 Peter Bankhead 13 John Wills 0000-0002-4347-5394 14 Paul Rees 0000-0002-7715-6914 15 Andrew Filby 16 George Merces 17 68202__33356__74e079d169674d4c88c8e7130f5286d2.pdf 68202.VoR.pdf 2025-01-16T12:25:34.6264025 Output 5214989 application/pdf Version of Record true © 2023 The Authors. This is an open access article under the terms of the Creative Commons Attribution License. true eng http://creativecommons.org/licenses/by/4.0/
title OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
spellingShingle OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
Paul Rees
title_short OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
title_full OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
title_fullStr OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
title_full_unstemmed OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
title_sort OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration
author_id_str_mv 537a2fe031a796a3bde99679ee8c24f5
author_id_fullname_str_mv 537a2fe031a796a3bde99679ee8c24f5_***_Paul Rees
author Paul Rees
author2 Bethany Hunter
Ioana Nicorescu
Emma Foster
David McDonald
Gillian Hulme
Andrew Fuller
Amanda Thomson
Thibaut Goldsborough
Catharien M. U. Hilkens
Joaquim Majo
Luke Milross
Andrew Fisher
Peter Bankhead
John Wills
Paul Rees
Andrew Filby
George Merces
format Journal article
container_title Cytometry Part A
container_volume 105
container_issue 1
container_start_page 36
publishDate 2024
institution Swansea University
issn 1552-4922
1552-4930
doi_str_mv 10.1002/cyto.a.24803
publisher Wiley
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Engineering and Applied Sciences - Biomedical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Engineering and Applied Sciences - Biomedical Engineering
document_store_str 1
active_str 0
description Analysis of imaging mass cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single-cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states, or spatial relationships compared to reference data from single-cell suspension technologies. To this end we have developed the “OPTimized Imaging Mass cytometry AnaLysis (OPTIMAL)” framework to benchmark any approaches for cell segmentation, parameter transformation, batch effect correction, data visualization/clustering, and spatial neighborhood analysis. Using a panel of 27 metal-tagged antibodies recognizing well-characterized phenotypic and functional markers to stain the same Formalin-Fixed Paraffin Embedded (FFPE) human tonsil sample tissue microarray over 12 temporally distinct batches we tested several cell segmentation models, a range of different arcsinh cofactor parameter transformation values, 5 different dimensionality reduction algorithms, and 2 clustering methods. Finally, we assessed the optimal approach for performing neighborhood analysis. We found that single-cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using “classical” bivariate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximized the statistical separation between negative and positive signal distributions and a simple Z-score normalization step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing phenograph in terms of cell type identification. We also found that neighborhood analysis was influenced by the method used for finding neighboring cells with a “disc” pixel expansion outperforming a “bounding box” approach combined with the need for filtering objects based on size and image-edge location. Importantly, OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output and allows for single-cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists.
published_date 2024-01-01T18:52:57Z
_version_ 1831847569499619328
score 11.058631

Similar Items