Högskolan i Skövde

his.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Comparative evaluation of DIABLO and NOLAS for multi-omics integration in cancer
University of Skövde, School of Bioscience.
2025 (English)Independent thesis Basic level (degree of Bachelor), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Multi-omics data integration has become a powerful approach for understanding complex diseases, such as cancer, and enables researchers to uncover biological mechanisms, identify biomarkers, and improve patient stratification. In this study, two supervised multi-omics integration methods were systematically compared, DIABLO (Data Integration Analysis for Biomarker Discovery using Latent variable approaches for Omics studies) and NOLAS (Noise-reduced Latent Structure), to predict breast cancer survival status (Alive/Deceased) using matched RNA-Seq, RPPA, and miRNA data from the TCGA-BRCA cohort. DIABLO, a multivariate block-integration method, was optimized for joint feature selection and classification, while NOLAS employed matrix factorization to reduce noise and extract latent structures. Both models were evaluated using stratified 50 train-50% test split, with performance assessed via AUC, F1-score and McNemar’s test (p< 0.001). Stability analysis (50 iterations with 80% subsampling) quantified feature selection consistency, followed by KEGG pathway enrichment to assess biological relevance. Results showed that DIABLO demonstrated slightly better classification performance than NOLAS in classification (AUC: 0.632 vs. 0.549), with McNemar’s test confirming significant differences (χ² = 71.63, p < 2.2×10⁻¹⁶). However, NOLAS showed higher feature stability in miRNA/RPPA layers (stability scores: 58.8% and 53.1% vs. DIABLO’s 51.28% and 38.46%). Enrichment analysis revealed DIABLO’s selected genes were linked to PI3K-Akt signaling, while NOLAS prioritized ECM-receptor interactions, highlighting distinct biological insights. These findings highlight the trade-offs between predictive accuracy and interpretability in multi-omics integration and offer insights relevant to decision making in precision oncology. The aim of this study is to systematically compare the supervised multi-omics integration methods DIABLO and NOLAS in predicting breast cancer survival. The comparison includes performance evaluation using multiple metrics, assessment of feature selection stability across RNA-Seq, RPPA, and miRNA data, and analysis of the biological relevance of stable features through KEGG pathway enrichment.

Place, publisher, year, edition, pages
2025. , p. 25
National Category
Medical Bioinformatics and Systems Biology
Identifiers
URN: urn:nbn:se:his:diva-25332OAI: oai:DiVA.org:his-25332DiVA, id: diva2:1973806
Subject / course
Bioinformatics
Educational program
Molekylär bioinformatik
Supervisors
Examiners
Available from: 2025-06-19 Created: 2025-06-19 Last updated: 2025-09-29Bibliographically approved

Open Access in DiVA

fulltext(754 kB)331 downloads
File information
File name FULLTEXT01.pdfFile size 754 kBChecksum SHA-512
674f14d547c4dc0fff1dcafc47c11601faaf080c25703aa4b5ee4a4a310118234018f0d7c993b3263052556d5706e8563475563369687d78171e0e8233169737
Type fulltextMimetype application/pdf

By organisation
School of Bioscience
Medical Bioinformatics and Systems Biology

Search outside of DiVA

GoogleGoogle Scholar
Total: 333 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 490 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf