Sample-specific perturbation of gene interactions identifies breast cancer subtypes

Abstract

Breast cancer is a highly heterogeneous disease, and there are many forms of categorization for breast cancer based on gene expression profiles. Gene expression profiles are variables and may show differences if measured at different time points or under different conditions. In contrast, biological networks are relatively stable over time and under different conditions. In this study, we used a gene interaction network from a new point of view to explore the subtypes of breast cancer based on individual-specific edge perturbations measured by relative gene expression value. Our study reveals that there are four breast cancer subtypes based on gene interaction perturbations at the individual level. The new network-based subtypes of breast cancer show strong heterogeneity in prognosis, somatic mutations, phenotypic changes and enriched pathways. The network-based subtypes are closely related to the PAM50 subtypes and immunohistochemistry index. This work helps us to better understand the heterogeneity and mechanisms of breast cancer from a network perspective.

Reassessment of Reliability and Reproducibility for Triple-Negative Breast Cancer Subtyping

“…Besides, certain features can be used to investigate specific questions, e.g., immune-related pathway signatures [ 10 , 11 ] or immune infiltration cell abundance [ 12 ] help to identify TNBC subtypes that may be responsive to immunotherapy. Moreover, gene expression profiles can be combined with other information, such as the protein interaction network, to develop new features and to obtain more stable and comprehensive clustering results [ 28 ].…”

A novel classification method for NSCLC based on the background interaction network and the edge-perturbation matrix

“…However, recent research on biological functional networks has discovered that heterogeneity can be avoided or minimized to a certain extent through the use of sequencing and bioinformatics analysis technology [ 10 – 14 ]. Thus, using the R language-based bioinformatics analysis technology, we designed the following research to re-evaluate the alterations in the NSCLC genome after excluding tumor heterogeneity: Based on edge perturbation [ 15 – 18 ], functional gene interaction networks were used to deduce the pathological environment of individual patients with NSCLC [ 19 – 22 ], and to identify cancer subtypes with the same or similar status, followed by a multi-dimensional and multi-omics comprehensive analysis for validation [ 23 – 26 ].…”

“…After successfully constructing a background interaction network and extracting gene expression ( Figure 1A and 1B ), it was found that the characteristic value of the cancer sample was significantly higher than that of the normal sample ( Figure 2A and 2B ). This indicated that the degree of edge perturbation in cancer samples was greater, implying that the degree of perturbation in the background network of cancer samples was significantly greater than in normal samples [ 15 ], providing reliable evidence for the subsequent use of EPM to reveal the heterogeneity of NSCLC samples.…”

“…The ReactomeFIPlugIn plug-in was used to download all genetic interaction pairs in the pathway and then merged into a large background network. The construction process of the edge disturbance matrix mainly included the following three steps [ 15 ].…”

“…Based on edge perturbation [ 15 – 18 ], functional gene interaction networks were used to deduce the pathological environment of NSCLC patients at the individual level [ 19 – 22 ], and to identify cancer subtypes with the same or similar status, followed by a multi-dimensional and multi-omics comprehensive analysis for validation [ 23 – 26 ].…”