# QNetDiff: a quantitative measurement of network rewiring

## TL;DR

This paper introduces QNetDiff, a new method to detect disease-associated bacteria by analyzing changes in bacterial network interactions, particularly in colorectal cancer.

## Contribution

The paper proposes a novel method to quantify network rewiring in bacterial communities for disease marker identification.

## Key findings

- QNetDiff identifies bacteria associated with colorectal cancer that are missed by traditional abundance comparison methods.
- The method is implemented as a publicly available tool for general use.
- Network rewiring analysis reveals new insights into bacterial interactions in disease states.

## Abstract

Bacteria in the human body, particularly in the large intestine, are known to be associated with various diseases. To identify disease-associated bacteria (markers), a typical method is to statistically compare the relative abundance of bacteria between healthy subjects and diseased patients. However, since bacteria do not necessarily cause diseases in isolation, it is also important to focus on the interactions and relationships among bacteria when examining their association with diseases. In fact, although there are common approaches to represent and analyze bacterial interaction relationships as networks, there are limited methods to find bacteria associated with diseases through network-driven analysis. In this paper, we focus on rewiring of the bacterial network and propose a new method for quantifying the rewiring. We then apply the proposed method to a group of colorectal cancer patients. We show that it can identify and detect bacteria that cannot be detected by conventional methods such as abundance comparison. Furthermore, the proposed method is implemented as a general-purpose tool and made available to the general public.