Academic Journal
A robust machine learning approach for breast cancer subtype classification using relative gene expression order representations.
| Τίτλος: | A robust machine learning approach for breast cancer subtype classification using relative gene expression order representations. |
|---|---|
| Συγγραφείς: | Uma Kandan S; Department of Electrical Engineering, College of Engineering, University of Sharjah, Sharjah, United Arab Emirates., Abul O; Department of Computer Science, College of Computing and Informatics, University of Sharjah, Sharjah, United Arab Emirates. Electronic address: oabul@sharjah.ac.ae. |
| Πηγή: | Computer methods and programs in biomedicine [Comput Methods Programs Biomed] 2026 Oct; Vol. 285, pp. 109510. Date of Electronic Publication: 2026 Jun 11. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Elsevier Scientific Publishers Country of Publication: Ireland NLM ID: 8506513 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1872-7565 (Electronic) Linking ISSN: 01692607 NLM ISO Abbreviation: Comput Methods Programs Biomed Subsets: MEDLINE |
| Imprint Name(s): | Publication: Limerick : Elsevier Scientific Publishers Original Publication: Amsterdam : Elsevier Science Publishers, c1984- |
| Ιατρικοί όροι (MeSH): | Breast Neoplasms*/genetics , Breast Neoplasms*/classification , Machine Learning* , Gene Expression Profiling*, Humans ; Female ; Classification Algorithms ; Gene Expression Regulation, Neoplastic ; Predictive Learning Models ; Algorithms |
| Περίληψη: | Background and Objective: Breast cancer subtype classification is critical for clinical decision-making. It informs prognostic assessment and guides personalized treatment planning. Machine learning has been widely applied to develop subtype classifiers based on gene expression profiles due to its ability to capture complex molecular patterns. However, technical variation in gene expression data presents a major challenge in building classifiers that maintain robust and generalizable performance across different platforms and patient cohorts. Methods: We propose a robust machine learning approach leveraging relative gene expression order representations for subtype classification. Two representations: rank- and word2vec embedding-based, were evaluated to capture biological variation across subtypes with minimal cross-sample dependence. These representations were derived from the within-sample relative expression order of PAM50 genes, a well-established 50-gene signature for defining breast cancer subtypes. Machine learning models were trained on these representations and systematically evaluated for robustness and clinical relevance across benchmark datasets. Results: Both rank- and word2vec embedding-based machine learning models demonstrated robust performance during cross-validation on the SCAN-B training set and on SCAN-B internal and semi-external test sets, achieving at least 91% precision and recall. On the fully external TCGA-BRCA dataset, the models accurately classified the majority of subtypes, with at least 95% accuracy. Predicted luminal subtypes also showed clinically meaningful stratification in survival analyses, and both representations effectively preserved biological variation, as reflected in subtype-wise sample segregation in PCA visualizations. In addition, word2vec-derived gene embeddings captured biologically meaningful co-occurring gene clusters. Conclusion: Ranks and word2vec embeddings derived from relative gene expression order enable machine learning models to robustly classify breast cancer subtypes. They address technical variability by reducing dependence on cohort-wide normalization, thereby supporting reliable subtype classification in diverse clinical settings. (Copyright © 2026 Elsevier B.V. All rights reserved.) |
| Competing Interests: | Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. |
| Contributed Indexing: | Keywords: Breast cancer subtype classification; Gene expression; Machine learning; Rank normalization; Robust classification; Word2vec embedding |
| Entry Date(s): | Date Created: 20260616 Date Completed: 20260806 Latest Revision: 20260806 |
| Update Code: | 20260807 |
| DOI: | 10.1016/j.cmpb.2026.109510 |
| PMID: | 42302593 |
| Βάση Δεδομένων: | MEDLINE |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://www.doi.org/10.1016/j.cmpb.2026.109510? Name: ScienceDirect (all content) (s7799221) Category: fullText Text: View record from ScienceDirect MouseOverText: View record from ScienceDirect – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:cmedm&genre=article&issn=18727565&ISBN=&volume=285&issue=&date=20261001&spage=109510&pages=109510&title=Computer methods and programs in biomedicine&atitle=A%20robust%20machine%20learning%20approach%20for%20breast%20cancer%20subtype%20classification%20using%20relative%20gene%20expression%20order%20representations.&aulast=Uma%20Kandan%20S&id=DOI:10.1016/j.cmpb.2026.109510 Name: Full Text Finder (for New FTF UI) (ns324271) Category: fullText Text: Full Text Finder MouseOverText: Full Text Finder |
|---|---|
| Header | DbId: cmedm DbLabel: MEDLINE An: 42302593 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: A robust machine learning approach for breast cancer subtype classification using relative gene expression order representations. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AU" term="%22Uma+Kandan+S%22">Uma Kandan S</searchLink>; Department of Electrical Engineering, College of Engineering, University of Sharjah, Sharjah, United Arab Emirates.<br /><searchLink fieldCode="AU" term="%22Abul+O%22">Abul O</searchLink>; Department of Computer Science, College of Computing and Informatics, University of Sharjah, Sharjah, United Arab Emirates. Electronic address: oabul@sharjah.ac.ae. – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%228506513%22">Computer methods and programs in biomedicine</searchLink> [Comput Methods Programs Biomed] 2026 Oct; Vol. 285, pp. 109510. <i>Date of Electronic Publication: </i>2026 Jun 11. – Name: TypePub Label: Publication Type Group: TypPub Data: Journal Article – Name: Language Label: Language Group: Lang Data: English – Name: TitleSource Label: Journal Info Group: Src Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22Elsevier+Scientific+Publishers%22">Elsevier Scientific Publishers </searchLink><i>Country of Publication: </i>Ireland <i>NLM ID: </i>8506513 <i>Publication Model: </i>Print-Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1872-7565 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2201692607%22">01692607 </searchLink><i>NLM ISO Abbreviation: </i>Comput Methods Programs Biomed <i>Subsets: </i>MEDLINE – Name: PublisherInfo Label: Imprint Name(s) Group: PubInfo Data: <i>Publication</i>: Limerick : Elsevier Scientific Publishers<br /><i>Original Publication</i>: Amsterdam : Elsevier Science Publishers, c1984- – Name: SubjectMESH Label: MeSH Terms Group: Su Data: <searchLink fieldCode="MM" term="%22Breast+Neoplasms%22">Breast Neoplasms*</searchLink>/<searchLink fieldCode="MM" term="%22Breast+Neoplasms+genetics%22">genetics</searchLink> <br /><searchLink fieldCode="MM" term="%22Breast+Neoplasms%22">Breast Neoplasms*</searchLink>/<searchLink fieldCode="MM" term="%22Breast+Neoplasms+classification%22">classification</searchLink> <br /><searchLink fieldCode="MM" term="%22Machine+Learning%22">Machine Learning*</searchLink> <br /><searchLink fieldCode="MM" term="%22Gene+Expression+Profiling%22">Gene Expression Profiling*</searchLink><br /><searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Female%22">Female</searchLink> ; <searchLink fieldCode="MH" term="%22Classification+Algorithms%22">Classification Algorithms</searchLink> ; <searchLink fieldCode="MH" term="%22Gene+Expression+Regulation%2C+Neoplastic%22">Gene Expression Regulation, Neoplastic</searchLink> ; <searchLink fieldCode="MH" term="%22Predictive+Learning+Models%22">Predictive Learning Models</searchLink> ; <searchLink fieldCode="MH" term="%22Algorithms%22">Algorithms</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Background and Objective: Breast cancer subtype classification is critical for clinical decision-making. It informs prognostic assessment and guides personalized treatment planning. Machine learning has been widely applied to develop subtype classifiers based on gene expression profiles due to its ability to capture complex molecular patterns. However, technical variation in gene expression data presents a major challenge in building classifiers that maintain robust and generalizable performance across different platforms and patient cohorts.<br />Methods: We propose a robust machine learning approach leveraging relative gene expression order representations for subtype classification. Two representations: rank- and word2vec embedding-based, were evaluated to capture biological variation across subtypes with minimal cross-sample dependence. These representations were derived from the within-sample relative expression order of PAM50 genes, a well-established 50-gene signature for defining breast cancer subtypes. Machine learning models were trained on these representations and systematically evaluated for robustness and clinical relevance across benchmark datasets.<br />Results: Both rank- and word2vec embedding-based machine learning models demonstrated robust performance during cross-validation on the SCAN-B training set and on SCAN-B internal and semi-external test sets, achieving at least 91% precision and recall. On the fully external TCGA-BRCA dataset, the models accurately classified the majority of subtypes, with at least 95% accuracy. Predicted luminal subtypes also showed clinically meaningful stratification in survival analyses, and both representations effectively preserved biological variation, as reflected in subtype-wise sample segregation in PCA visualizations. In addition, word2vec-derived gene embeddings captured biologically meaningful co-occurring gene clusters.<br />Conclusion: Ranks and word2vec embeddings derived from relative gene expression order enable machine learning models to robustly classify breast cancer subtypes. They address technical variability by reducing dependence on cohort-wide normalization, thereby supporting reliable subtype classification in diverse clinical settings.<br /> (Copyright © 2026 Elsevier B.V. All rights reserved.) – Name: Abstract Label: Competing Interests Group: Ab Data: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. – Name: SubjectMinor Label: Contributed Indexing Group: Data: <i>Keywords: </i>Breast cancer subtype classification; Gene expression; Machine learning; Rank normalization; Robust classification; Word2vec embedding – Name: DateEntry Label: Entry Date(s) Group: Date Data: <i>Date Created: </i>20260616 <i>Date Completed: </i>20260806 <i>Latest Revision: </i>20260806 – Name: DateUpdate Label: Update Code Group: Date Data: 20260807 – Name: DOI Label: DOI Group: ID Data: 10.1016/j.cmpb.2026.109510 – Name: AN Label: PMID Group: ID Data: 42302593 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42302593 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1016/j.cmpb.2026.109510 Languages: – Code: eng Text: English PhysicalDescription: Pagination: StartPage: 109510 Subjects: – SubjectFull: Humans Type: general – SubjectFull: Female Type: general – SubjectFull: Classification Algorithms Type: general – SubjectFull: Gene Expression Regulation, Neoplastic Type: general – SubjectFull: Predictive Learning Models Type: general – SubjectFull: Algorithms Type: general – SubjectFull: Breast Neoplasms genetics Type: general – SubjectFull: Breast Neoplasms classification Type: general – SubjectFull: Machine Learning Type: general – SubjectFull: Gene Expression Profiling Type: general Titles: – TitleFull: A robust machine learning approach for breast cancer subtype classification using relative gene expression order representations. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Uma Kandan S – PersonEntity: Name: NameFull: Abul O IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 10 Text: 2026 Oct Type: published Y: 2026 Identifiers: – Type: issn-electronic Value: 1872-7565 Numbering: – Type: volume Value: 285 Titles: – TitleFull: Computer methods and programs in biomedicine Type: main |
| ResultId | 1 |