Predicting Host Phenotype Based on Gut Microbiome Using a Convolutional Neural Network Approach

Reiman, Derek; Farhat, Ali M.; Dai, Yang

doi:10.1007/978-1-0716-0826-5_12

Cited by 8 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Investigating the functional phenotypes from microbial samples is essential in understanding the impact of gut microbiota alterations on interaction and homeostasis with the host ( 22 ). To further explore the functional differences of gut microbiota between the three groups, microbiome phenotype prediction was carried out using Bugbase ( https://bugbase.cs.umn.edu ).…”

Section: Resultsmentioning

confidence: 99%

Potential risk of tamoxifen: gut microbiota and inflammation in mice with breast cancer

Li¹,

Gao

Chen

et al. 2023

Front. Oncol.

View full text Add to dashboard Cite

ObjectiveTamoxifen is an effective anti-tumor medicine, but evidence has been provided on tamoxifen-related inflammation as well as its impact on gut microbiota. In this study, we aimed to investigate tamoxifen-induced gut microbiota and inflammation alteration.MethodsWe established a BC xenograft mouse model using the MCF-7 cell line. 16S rRNA gene sequencing was used to investigate gut microbiota. qRT–PCR, western blotting, and cytometric bead array were used to investigate inflammation-related biomarkers. Various bioinformatic approaches were used to analyze the data.ResultsSignificant differences in gut microbial composition, characteristic taxa, and microbiome phenotype prediction were observed between control, model, and tamoxifen-treated mice. Furthermore, protein expression of IL-6 and TLR5 was up-regulated in tamoxifen-treated mice, while the mRNA of Tlr5 and Il-6, as well as protein expression of IL-6 and TLR5 in the model group, were down-regulated in the colon. The concentration of IFN-γ, IL-6, and IL12P70 in serum was up-regulated in tamoxifen-treated mice. Moreover, correlation-based clustering analysis demonstrated that inflammation-negatively correlated taxa, including Lachnospiraceae-UCG-006 and Anaerotruncus, were enriched in the model group, while inflammation-positively correlated taxa, including Prevotellaceae_UCG_001 and Akkermansia, were enriched in the tamoxifen-treated group. Finally, colon histologic damage was observed in tamoxifen-treated mice.ConclusionTamoxifen treatment significantly altered gut microbiota and increased inflammation in the breast cancer xenograft mice model. This may be related to tamoxifen-induced intestinal epithelial barrier damage and TLR5 up-regulation.

show abstract

Section: Resultsmentioning

confidence: 99%

Potential risk of tamoxifen: gut microbiota and inflammation in mice with breast cancer

Li¹,

Gao

Chen

et al. 2023

Front. Oncol.

View full text Add to dashboard Cite

show abstract

“…Thus, the presented approach provides a simple and accurate classification. Problems with custom CNNs [25][26][27][28][29] include shallow depth architectures with few layers or classifier overfitting. ResNets did not achieve success because ImageNet [32] pretrained weights were used [27], or the data visualization as a bar chart of microbiome abundances [41] was simply not suitable for the network.…”

Section: Discussionmentioning

confidence: 99%

“…Gene expression and microbiome sequence data have been visualized using different methods, and CNNs have been used to classify phenotypes and diseases. Reiman et al generated phylogenetic and taxonomic trees of microbiome data and transformed them into matrices [25,26]. These could be classified as images to predict phenotypes of origin such as skin, mouth, and gut.…”

Section: Introductionmentioning

confidence: 99%

Classification of Microbiome Data from Type 2 Diabetes Mellitus Individuals with Deep Learning Image Recognition

Pfeil

Siptroth

Pospisil

et al. 2023

BDCC

View full text Add to dashboard Cite

Microbiomic analysis of human gut samples is a beneficial tool to examine the general well-being and various health conditions. The balance of the intestinal flora is important to prevent chronic gut infections and adiposity, as well as pathological alterations connected to various diseases. The evaluation of microbiome data based on next-generation sequencing (NGS) is complex and their interpretation is often challenging and can be ambiguous. Therefore, we developed an innovative approach for the examination and classification of microbiomic data into healthy and diseased by visualizing the data as a radial heatmap in order to apply deep learning (DL) image classification. The differentiation between 674 healthy and 272 type 2 diabetes mellitus (T2D) samples was chosen as a proof of concept. The residual network with 50 layers (ResNet-50) image classification model was trained and optimized, providing discrimination with 96% accuracy. Samples from healthy persons were detected with a specificity of 97% and those from T2D individuals with a sensitivity of 92%. Image classification using DL of NGS microbiome data enables precise discrimination between healthy and diabetic individuals. In the future, this tool could enable classification of different diseases and imbalances of the gut microbiome and their causative genera.

show abstract

“…One approach is to capture the spatial information from the relative abundance profiles of microbial taxa; transform the data into taxonomic trees for further embedding into a 2‐D matrix. Reiman's proposed CNN architecture achieved a 99.47% of accuracy on disease prediction from data collected by 16s RNA sequencing 33,34 . Another proposed method is to convert the multi‐omics data into graphical representations, such as heatmap.…”

Section: Artificial Intelligence Machine Learning and Deep Learningmentioning

confidence: 99%

“…Reiman's proposed CNN architecture achieved a 99.47% of accuracy on disease prediction from data collected by 16s RNA sequencing. 33,34 Another proposed method is to convert the multi-omics data into graphical representations, such as heatmap. Heatmap is widely used in microbiome studies to provide intuitive visual representations of the numerical data.…”

Section: Artificial Intelligence Machine Learning and Deep Learningmentioning

confidence: 99%

Machine learning on microbiome research in gastrointestinal cancer

Cheung

2021

J of Gastro and Hepatol

View full text Add to dashboard Cite

Gastrointestinal cancer maintains the highest incidence and mortality rate among all cancers globally. In addition to genetic causes, it has been reported that individuals' diet and composition of the gastrointestinal microbiome have profound impacts on gastrointestinal cancer development. Microbiome research has risen in popularity to provide alternative insights into cancer development and potential therapeutic effect. However, there is a lack of an effective analytical tool to comprehend the massive amount of data generated from high‐throughput sequencing methods. Artificial intelligence is another rapidly developing field that has strong application potential in microbiome research. Deep learning and machine learning are two subfields under the umbrella of artificial intelligence. Here we discuss the current approaches to study the gut microbiome, as well as the applications and challenges of implementing artificial intelligence in microbiome research.

show abstract

Predicting Host Phenotype Based on Gut Microbiome Using a Convolutional Neural Network Approach

Cited by 8 publications

References 23 publications

Potential risk of tamoxifen: gut microbiota and inflammation in mice with breast cancer

Potential risk of tamoxifen: gut microbiota and inflammation in mice with breast cancer

Classification of Microbiome Data from Type 2 Diabetes Mellitus Individuals with Deep Learning Image Recognition

Machine learning on microbiome research in gastrointestinal cancer

Contact Info

Product

Resources

About