Measurement of structural change in co-expression networks from cancer gene expression data

Qianran Li, Dario Ghersi, Ishwor Thapa, Ling Zhang, Hesham Ali, Kate Cooper

Research output: Contribution to journalArticlepeer-review


Profiling progression and development of cancer is an important step for informing clinical decisions in the diagnosis, prognosis and treatment of cancer. The hallmarks of cancer progression are well established; for example, many studies have concluded that surveillance of gene expression relationships during cancer progression would inform diagnostic and treatment decisions. Differential network analysis offers a systems-level insight into cancer progression from a high-level point of view, and once further understood, could become a transformative approach for measuring cancer progression. In this work, we investigate an approach to measure pairwise change in network topology between cancer stages for four cancers, including thyroid carcinoma, colon and rectum adenocarcinoma, stomach adenocarcinoma, and kidney renal papillary cell carcinoma. We use a network-based approach to describe systems-level views of how a network model changes over the course of a four-stage disease progression in these cancers and examine how mutation rate corresponds to network structure. Lastly, we present a case study in comparing primary versus metastatic tumour network structure. The results of this study demonstrate the applicability of such an approach and provide insights into next steps that are needed for differential network comparison in cancer.

Original languageEnglish (US)
Pages (from-to)291-305
Number of pages15
JournalInternational Journal of Data Mining and Bioinformatics
Issue number4
StatePublished - 2020


  • Correlation network
  • Gene expression
  • Jaccard similarity
  • Mutation rate
  • The cancer genome atlas

ASJC Scopus subject areas

  • Information Systems
  • Biochemistry, Genetics and Molecular Biology(all)
  • Library and Information Sciences


Dive into the research topics of 'Measurement of structural change in co-expression networks from cancer gene expression data'. Together they form a unique fingerprint.

Cite this