Visualizing Microbiome

Our bodies are home to a diverse array of microorganisms that play a critical role in our overall health and well-being. Understanding the intricacies of these microbial communities is crucial for diagnosing and treating a range of diseases, from irritable bowel syndrome to cancers and obesity. The challenge lies in processing and interpreting the vast amounts of data generated by these studies. That's why the development of new data visualization tools is essential for unlocking the full potential of microbiome research. 

The latest methods for visualizing and analyzing microbial population structures include freely accessible web-based tools such as Visualization and Analysis of Microbial Population Structures (VAMPS), MicrobiomeAnalyst, Mian, gcMeta, Microbiome Toolbox, and R-based Genepiper and MANTA.

The gut microbiome changes rapidly under the influence of different factors such as age, dietary changes or medications to name just a few. 

Exploring and understanding changes in the microbiome in relation to different factors such as time or age, changes in the environment and diet, as well as medications, are of great interest especially as these relate to health conditions. 

One of the most useful tools to analyze and understand such changes is Microbiome Toolbox. This web-based tool helps to depict microbiome change as trajectories under different conditions such as time, diet changes or perturbations. Microbiome Toolbox has implemented several complex algorithms, but it also provides a rich variety of plots for easy visual comprehension and reporting. A machine-learning-based approach is used to derive a microbiome trajectory. 

Similar to MicrobiomeAnalyst, an interactive web-based data discovery platform Mian, makes use of the common input file formats (BIOM, CSV/TSV-formatted OTU/ASV tables) generated from Mothur, QIIME, and DADA2. The users can visualize their alpha and beta diversity metrics data using stacked bars, heatmaps, box, donut and scatter plots, PCoA, and NMDS. Unlike MicrobiomeAnalyst, which uses LEfSe for feature selection, Mian uses recursive feature elimination, Fisher’s exact test, and Boruta, which selects the OTUs/ASVs or taxonomic groups that are applied on a random forest classifier and are ideal for selecting all of the groups that are relevant for discriminating between populations, in contrast to finding the non-redundant ones. Mian offers the use of machine learning tools to assess the discriminative performance of the taxonomic groups selected through a feature selection tool. Tools, including Fisher's exact test, Boruta feature selection, alpha and beta diversity, and random forest and deep neural network classifiers, facilitate open-ended data exploration and hypothesis generation on microbial datasets. 


REFERENCES

Banjac J, Sprenger N, Dogra SK. Microbiome Toolbox: methodological approaches to derive and visualize microbiome trajectories. Bioinformatics. 2023 Jan 1;39(1):btac781. doi: 10.1093/bioinformatics/btac781. PMID: 36469345; PMCID: PMC9825749.

Jin BT, Xu F, Ng RT, Hogg JC. Mian: interactive web-based microbiome data table visualization and machine learning platform. Bioinformatics. 2022 Jan 27;38(4):1176-1178. doi: 10.1093/bioinformatics/btab754. PMID: 34788784.

Ibal JC, Park YJ, Park MK, Lee J, Kim MC, Shin JH. Review of the Current State of Freely Accessible Web Tools for the Analysis of 16S rRNA Sequencing of the Gut Microbiome. Int J Mol Sci. 2022 Sep 17;23(18):10865. doi: 10.3390/ijms231810865. PMID: 36142775; PMCID: PMC9501225.

JelenaBanjac/microbiome-toolbox: Early Life Microbiome Toolbox (github.com)

tbj128/mian: Mian is a platform for 16S rRNA OTU table visualization, data analysis, and feature selection (github.com)

Mian: 16S rRNA Operational Taxonomic Unit (OTU) data analysis, visualization, and feature selection tools (miandata.org)

Comments

Popular posts from this blog

Microbes of Autobrewery Syndrome

Lactobacillus & Bifidobacterium

Genetically Engineered Probiotics