Parenclitic network analysis of methylation data for cancer identification

TL;DR

This paper introduces a network-based machine learning approach using parenclictic networks to classify cancer types from DNA methylation data, achieving high accuracy with reduced complexity and revealing biological insights.

Contribution

It presents a novel application of parenclictic network analysis for cancer classification, significantly reducing feature dimensionality and uncovering network properties related to cancer.

Findings

High classification accuracy (93-99%) with only 12 network indices.

Parenclictic networks are scale-free in healthy subjects.

Deviations from scale-free behavior in cancer networks.

Abstract

We make use of ideas from the theory of complex networks to implement a machine learning classification of human DNA methylation data, that carry signatures of cancer development. The data were obtained from patients with various kinds of cancers and represented as parenclictic networks, wherein nodes correspond to genes, and edges are weighted according to pairwise variation from control group subjects. We demonstrate that for the $10$ types of cancer under study, it is possible to obtain a high performance of binary classification between cancer-positive and negative samples based on network measures. Remarkably, an accuracy as high as $93-99\%$ is achieved with only $12$ network topology indices, in a dramatic reduction of complexity from the original $15295$ gene methylation levels. Moreover, it was found that the parenclictic networks are scale-free in cancer-negative subjects, and deviate from the power-law node degree distribution in cancer. The node centrality ranking and arising modular structure could provide insights into the systems biology of cancer.