• DriverSub-SVM: a machine learn...

*Result*: DriverSub-SVM: a machine learning approach for cancer subtype classification by integrating patient-specific and global driver genes.

Title:
DriverSub-SVM: a machine learning approach for cancer subtype classification by integrating patient-specific and global driver genes.
Authors:
Song J; School of Information, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China. zz2144@ynufe.edu.cn., Gong Y; School of Information, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China., Song Z; School of Information, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China. zz2145@ynufe.edu.cn.; Yunnan Key Laboratory of Service Computing, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China. zz2145@ynufe.edu.cn., Xu X; School of Information, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China., Qian K; School of Information, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China., Liu Y; School of Information, Yunnan University of Finance and Economics, Kunming, 650221, Yunnan, People's Republic of China. liuyb@ynufe.edu.cn.
Source:
BMC bioinformatics [BMC Bioinformatics] 2025 Dec 29; Vol. 26 (1), pp. 297. Date of Electronic Publication: 2025 Dec 29.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: BioMed Central Country of Publication: England NLM ID: 100965194 Publication Model: Electronic Cited Medium: Internet ISSN: 1471-2105 (Electronic) Linking ISSN: 14712105 NLM ISO Abbreviation: BMC Bioinformatics Subsets: MEDLINE
Imprint Name(s):
Original Publication: [London] : BioMed Central, 2000-
MeSH Terms:
Neoplasms*/genetics , Neoplasms*/classification , Support Vector Machine* , Machine Learning*, Humans ; Bayes Theorem ; Algorithms
References:
Cell Biosci. 2020 Mar 10;10:31. (PMID: 32175074)
Infect Agent Cancer. 2023 Feb 26;18(1):13. (PMID: 36843070)
Med Microbiol Immunol. 2023 Aug;212(4):263-270. (PMID: 37222763)
Cancers (Basel). 2022 Oct 29;14(21):. (PMID: 36358747)
Brief Bioinform. 2025 Jul 2;26(4):. (PMID: 40874818)
Mol Ther Oncolytics. 2017 Oct 24;8:1-13. (PMID: 29234727)
J Clin Oncol. 2009 Mar 10;27(8):1160-7. (PMID: 19204204)
Hum Pathol. 2019 Feb;84:192-201. (PMID: 30342055)
Oncogene. 2000 Dec 11;19(53):6102-14. (PMID: 11156523)
Nature. 2013 Jul 11;499(7457):214-218. (PMID: 23770567)
BMC Bioinformatics. 2019 May 14;20(1):238. (PMID: 31088372)
J Natl Cancer Inst. 2003 Apr 16;95(8):625-7. (PMID: 12697856)
Comput Struct Biotechnol J. 2021 Jan 22;19:949-960. (PMID: 33613862)
Int J Mol Sci. 2019 Oct 09;20(20):. (PMID: 31600974)
Int J Mol Sci. 2022 Oct 20;23(20):. (PMID: 36293435)
PLoS Comput Biol. 2024 May 10;20(5):e1012113. (PMID: 38728362)
Cell Oncol (Dordr). 2023 Oct;46(5):1213-1234. (PMID: 37166744)
Comput Methods Programs Biomed. 2017 Apr;141:27-34. (PMID: 28241965)
Technol Health Care. 2022;30(3):565-578. (PMID: 34397436)
Oncogene. 2021 Jan;40(4):777-790. (PMID: 33262463)
Genes (Basel). 2020 Aug 04;11(8):. (PMID: 32759821)
J Biomed Inform. 2020 Jul;107:103466. (PMID: 32525020)
Science. 2013 Mar 29;339(6127):1546-58. (PMID: 23539594)
Nucleic Acids Res. 2022 Jan 7;50(D1):D701-D709. (PMID: 34634810)
Turk J Biol. 2023 Dec 15;47(6):406-412. (PMID: 38681775)
Cancers (Basel). 2019 Nov 05;11(11):. (PMID: 31694222)
Cancer Cell Int. 2023 Nov 26;23(1):296. (PMID: 38008753)
Br J Cancer. 2022 Mar;126(5):797-803. (PMID: 34949788)
Cell Commun Signal. 2023 Sep 15;21(1):232. (PMID: 37715239)
Nature. 2014 Sep 11;513(7517):202-9. (PMID: 25079317)
Cancer Epidemiol Biomarkers Prev. 2019 Apr;28(4):643-649. (PMID: 30541751)
Cancer Res. 2005 Nov 15;65(22):10199-207. (PMID: 16288007)
Int J Oncol. 2015 Apr;46(4):1421-34. (PMID: 25633561)
Acta Pharmacol Sin. 2015 Oct;36(10):1170-6. (PMID: 26364801)
Genomics. 2008 Dec;92(6):400-3. (PMID: 18565726)
Zhonghua Bing Li Xue Za Zhi. 2022 Jun 8;51(6):536-541. (PMID: 35673726)
J Oncol. 2019 Jun 02;2019:5952836. (PMID: 31275382)
IEEE/ACM Trans Comput Biol Bioinform. 2022 Mar-Apr;19(2):1193-1202. (PMID: 32750893)
Endocrinol Metab (Seoul). 2021 Feb;36(1):96-105. (PMID: 33677931)
iScience. 2022 Aug 27;25(9):105023. (PMID: 36105596)
Bioinformatics. 2018 Jul 1;34(13):i404-i411. (PMID: 29950003)
Eur J Cancer. 2021 Mar;145:92-108. (PMID: 33429148)
NPJ Breast Cancer. 2021 Oct 12;7(1):136. (PMID: 34642313)
J Med Syst. 2019 Jun 17;43(8):235. (PMID: 31209677)
J Gastrointest Surg. 2021 Sep;25(9):2231-2241. (PMID: 33420656)
Mol Cancer. 2023 Aug 18;22(1):138. (PMID: 37596643)
Int J Mol Sci. 2020 May 19;21(10):. (PMID: 32438653)
Cancer Res. 2016 Jul 1;76(13):3989-4001. (PMID: 27197157)
Contemp Oncol (Pozn). 2015;19(1A):A68-77. (PMID: 25691825)
Sci Rep. 2024 Aug 1;14(1):17795. (PMID: 39090342)
Curr Cancer Drug Targets. 2008 Dec;8(8):733-40. (PMID: 19075596)
Front Genet. 2019 Mar 28;10:236. (PMID: 30984238)
Hum Pathol. 2020 Nov;105:53-66. (PMID: 32971129)
Nature. 2013 Sep 19;501(7467):355-64. (PMID: 24048068)
Genome Biol. 2010;11(5):R53. (PMID: 20482850)
Sci Rep. 2024 May 31;14(1):12542. (PMID: 38822093)
PLoS One. 2017 Sep 5;12(9):e0184129. (PMID: 28873455)
Nat Genet. 2023 Apr;55(4):581-594. (PMID: 36914835)
Nature. 2013 Oct 17;502(7471):333-339. (PMID: 24132290)
Med J Islam Repub Iran. 2023 Sep 18;37:101. (PMID: 38021380)
Brief Bioinform. 2022 May 13;23(3):. (PMID: 35437603)
Nucleic Acids Res. 2019 Jul 2;47(W1):W587-W593. (PMID: 31114887)
Br J Surg. 2022 Feb 24;109(3):291-297. (PMID: 35179206)
Brief Bioinform. 2024 Nov 22;26(1):. (PMID: 39879387)
Cancer Sci. 2021 Jun;112(6):2097-2117. (PMID: 33811715)
JAMA Otolaryngol Head Neck Surg. 2023 Jan 1;149(1):79-86. (PMID: 36454559)
Cancers (Basel). 2023 Jan 24;15(3):. (PMID: 36765665)
Lab Invest. 2022 Dec;102(12):1314-1322. (PMID: 35851857)
IEEE/ACM Trans Comput Biol Bioinform. 2022 Sep-Oct;19(5):2863-2872. (PMID: 34415837)
J Natl Cancer Inst. 2015 Nov 23;108(4):. (PMID: 26598514)
Bioinformatics. 2007 Feb 15;23(4):401-7. (PMID: 17182697)
Cancer Cell Int. 2022 Aug 22;22(1):263. (PMID: 35996174)
Cancers (Basel). 2024 May 30;16(11):. (PMID: 38893181)
Front Genet. 2021 Feb 24;12:632620. (PMID: 33719342)
Transl Oncol. 2017 Dec;10(6):956-975. (PMID: 29078205)
Genome Biol. 2022 Jan 26;23(1):35. (PMID: 35078504)
Grant Information:
62202415 National Natural Science Foundation of China; 62161051 National Natural Science Foundation of China; 62362064 National Natural Science Foundation of China; 202201AU070115 Yunnan Fundamental Research Projects; 202201AU070116 Yunnan Fundamental Research Projects; 2024D44 Talent Introduction Project of Yunnan University of Finance and Economics; 2025YUFEYC101 the Graduate Innovation Fund Project Yunnan University of Finance and Economics
Contributed Indexing:
Keywords: Cancer subtype classification; Concensus gene selection; Driver gene identification; Interpretable machine learning; Multiclass SVM; Personalized ranking
Entry Date(s):
Date Created: 20251230 Date Completed: 20251230 Latest Revision: 20260101
Update Code:
20260130
PubMed Central ID:
PMC12751776
DOI:
10.1186/s12859-025-06318-7
PMID:
41462070
Database:
MEDLINE

*Further Information*

*Background: Cancer's complexity and heterogeneity pose significant challenges for personalized treatment. Accurate classification of patients into molecular subtypes is critical for targeted therapy and improved outcomes. However, existing methods often fail to simultaneously capture inter-patient heterogeneity and shared molecular patterns in driver gene profiles.
Results: To address this limitation, we propose DriverSub-SVM, a novel framework for interpretable cancer subtype classification that integrates patient-specific and cohort-wide driver gene information. Our method first models the bidirectional influence between mutated and dysregulated genes via a random walk on a functional interaction network. It then applies Bayesian Personalized Ranking (BPR) to infer personalized driver gene rankings for each patient. These rankings are aggregated into a consensus driver gene set using the Condorcet. Subsequently, a One-Against-One Multiclass Support Vector Machine (OAO-MSVM) classifies patients based on their gene-level profiles. Evaluated on multiple TCGA datasets, DriverSub-SVM outperformed four state-of-the-art methods, achieving higher accuracy and identifying clinically relevant genes associated with prognosis and therapeutic response.
Conclusion: DriverSub-SVM offers an effective and interpretable approach for cancer subtype classification by bridging individual heterogeneity and population-level patterns. It enhances understanding of tumor biology and holds promise for precision oncology and biomarker discovery. The source code is available at https://github.com/sjunrong/DriverSub-SVM .
(© 2025. The Author(s).)*

*Declarations. Ethics approval and consent to participate: Not applicable. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests. Clinical trial number: Not applicable.*