Source-linked AI summary
ggpicrust2: an R package for PICRUSt2 predicted functional profile analysis and visualization
Chen Yang, Jiahao Mai, Xuan Cao, Aaron Burberry, Fabio Cominelli, Liangliang Zhang
TL;DR
PICRUSt2 predicts microbial pathway abundances from 16S data, but tools for differential analysis and visualization remain limited and lack consensus. ggpicrust2 integrates advanced differential-abundance methods, pathway conversion and annotation, and visualization; in a mouse application, it identified significant pathway differences between environments.
Problem
Marker-gene sequencing lacks direct functional information, and no consensus methodology exists for differential analysis and visualization of PICRUSt2 pathway abundances.
Method
ggpicrust2 integrates advanced differential-abundance methods with KO-to-KEGG pathway conversion, functional annotation, pathway classification, and visualization.
Results
In C9orf72 loss-of-function mice, ggpicrust2 identified statistically significant KEGG pathway differences between pro-survival and pro-inflammatory environments across transplantation groups.
Takeaways & Limitations
ggpicrust2 provides an integrated workflow for analyzing and visualizing PICRUSt2-predicted functional profiles for users with differing levels of expertise.
Abstract
from arXiv · showhide
Microbiome research is now moving beyond the compositional analysis of microbial taxa in a sample. Increasing evidence from large human microbiome studies suggests that functional consequences of changes in the intestinal microbiome may provide more power for studying their impact on inflammation and immune responses. Although 16S rRNA analysis is one of the most popular and a cost-effective method to profile the microbial compositions, marker-gene sequencing cannot provide direct information about the functional genes that are present in the genomes of community members. Bioinformatic tools have been developed to predict microbiome function with 16S rRNA gene data. Among them, PICRUSt2 has become one of the most popular functional profile prediction tools, which generates community-wide pathway abundances. However, no state-of-art inference tools are available to test the differences in pathway abundances between comparison groups. We have developed ggpicrust2, an R package, to do extensive differential abundance(DA) analyses and provide publishable visualization to highlight the signals.
1 Introduction
Marker-gene sequencing does not directly reveal microbial functional composition, prompting development of methods that predict functions from 16S rRNA data. PICRUSt2 is favored for generating pathway abundances, but appropriate differential-abundance and visualization methods remain unsettled.
- Marker-gene sequencing cannot provide information about the functional composition of sample communities.
- Methods including PICRUSt2, Tax4Fun2, MicFunPred, and PICRUSt predict microbial functions from 16S rRNA sequences.Their accuracy and applicability depend on the research question and microbial community characteristics.
- PICRUSt2 has emerged as a favored tool because it generates comprehensive pathway abundances within microbial communities.
- The optimal methodology for inferring and visualizing PICRUSt2 functional-abundance outputs, including differential-abundance testing between groups, remains unresolved.Selecting an appropriate differential-abundance approach is considered critical and important.
2 ggpicrust2 R package
ggpicrust2 supports PICRUSt2 functional-profile analysis by converting and annotating pathway data, integrating differential-abundance methods, and generating visualizations. In a mouse microbiome example, LinDA identified significantly different KEGG pathways between pro-survival and pro-inflammatory environments.
- Core functionality: ggpicrust2 converts PICRUSt2 KO abundance tables into KEGG pathway abundance tables and supports pathway annotation and KEGG pathway classification.The ko2kegg_abundance() function addresses conversion that PICRUSt2 alone cannot perform.
- Differential abundance: pathway_daa() integrates ALDEx2, DESeq2, Maaslin2, LinDA, edgeR, limma voom, metagenomeSeq, and lefser, excluding ANCOM and ANCOMBC.The package plans to incorporate additional methods supported by simulation comparisons.
- Visualization: ggpicrust2 provides bar, error-bar, PCA, and heatmap visualizations, including relative-abundance differences, log2 fold changes, p-values, dimensional reduction, and data patterns.pathway_errorbar() displays group differences with differential-abundance statistics, while pathway_pca() and pathway_heatmap() summarize multivariate patterns.
- Workflow integration: ggpicrust() integrates pathway_daa(), pathway_annotation(), pathway_errorbar(), and ko2kegg_abundance() into a workflow supporting novice use and modular customization by professional analysts.A detailed user manual provides installation instructions and explanations of the package’s main features.
- Application example: LinDA identified statistically significant KEGG pathway differences between pro-survival and pro-inflammatory environments across C9orf72 loss-of-function mice with or without fecal transplantation.The highlighted pathways were ko05016, associated with Huntington’s disease, and ko05012, associated with Parkinson’s disease.
3 Conclusion
ggpicrust2 is an R package for advanced differential abundance analysis and visualization of PICRUSt2-predicted functional profiles. Its integrated design addresses limitations of existing tools while serving both professionals and beginners.
- 3 Conclusion: ggpicrust2 is available through CRAN and GitHub for advanced differential abundance analysis and visualization of PICRUSt2-predicted functional profiles.The package was developed explicitly for PICRUSt2-predicted functional profiles.
- 3 Conclusion: The package addresses existing tools’ limitations in both analytical methods and visualization.
- 3 Conclusion: Its integrated and distributed design supports the needs of both professionals and beginners.