A neural network‐based framework to understand the type 2 diabetes‐related alteration of the human gut microbiome

Abstract

The identification of microbial markers adequate to delineate the disease‐related microbiome alterations from the complex human gut microbiota is of great interest. Here, we develop a framework combining neural network (NN) and random forest, resulting in 40 marker species and 90 marker genes identified from the metagenomic data set (185 healthy and 183 type 2 diabetes [T2D] samples), respectively. In terms of these markers, the NN model obtained higher accuracy in classifying the T2D‐related samples than other methods; the interaction network analyses identified the key species and functional modules; the regression analysis determined that fasting blood glucose is the most significant factor (p < 0.05) in the T2D‐related alteration of the human gut microbiome. We also observed that those marker species varied little across the case and control samples greatly shift in the different stages of the T2D development, suggestive of their important roles in the T2D‐related microbiome alteration. Our study provides a new way of identifying the disease‐related biomarkers and analyzing the role they may play in the development of the disease.

Deep learning methods in metagenomics: a review

“…However, methods like Principal Component Analysis have shown satisfying results when applied on this data, leading to a slight improvement in prediction. The other way around, ML methods like [48] and [146] aim to extract top decisive features or markers for disease prediction to understand key roles played by these features in the apparition of a disease.…”

MetDIT: Transforming and Analyzing Clinical Metabolomics Data with Convolutional Neural Networks

“…Recent studies − have attempted to incorporate multilayer perceptions (MLP) or recurrent neural networks (RNN) into omics data analysis, primarily on RNA sequencing data. For instance, Guo et al proposed a combined framework with an MLP and RF to understand the human microbiome for identifying the T2D-related microbial markers. It trained an MLP model with biomarkers selected by RF and achieved better performance than other machine learning methods.…”

“…For instance, Guo et al 26 proposed a combined framework with an MLP and RF to understand the human microbiome for identifying the T2D-related microbial markers. It trained an MLP model with biomarkers selected by RF and achieved better performance than other machine learning methods.…”

“…Since the launch of AlexNet, convolutional neural networks (CNN) have been the dominant solution for computer vision tasks, such as image classification, ,, object detection, , and semantic segmentation . Recent studies − have attempted to incorporate multilayer perceptions (MLP) or recurrent neural networks (RNN) into omics data analysis, primarily on RNA sequencing data. For instance, Guo et al proposed a combined framework with an MLP and RF to understand the human microbiome for identifying the T2D-related microbial markers.…”

Longevity-Associated Core Gut Microbiota Mining and Effect of Mediated Probiotic Combinations on Aging Mice: Case Study of a Long-Lived Population in Guangxi, China

“…However, using more than one method for analysis may be more susceptible to noisy samples. Some researchers have found that by combining RF with other algorithms, RF can use the mechanism of multiple tree voting to cancel out some noise [ 20 ]. However, studies have yet to report combining RF and WGCNA to better analyse and mine the intestinal microflora.…”