The Microbial Analysis module focuses on profiling microbiome composition across samples, producing standardized abundance tables and visualization-ready outputs for downstream statistical and machine learning analysis.
What This Module Does
Microbial Analysis is designed to provide a comprehensive overview of microbial community composition across samples, cohorts, or experimental conditions.
It answers questions such as:
- Which microbial taxa are present?
- How do communities differ across samples or groups?
- Which taxa dominate or vary across conditions?
The module produces both human-interpretable visualizations and machine-ready tables for downstream analysis.
Input Data
Required Inputs
- FASTQ files
- Single-end or paired-end sequencing reads
- Compatible with standard Illumina workflows
Optional Inputs
- Metadata table (CSV / TSV)
- Sample identifiers (required for grouping)
- Group or condition labels
- Covariates (batch, timepoint, treatment, etc.)
Well-structured metadata enables stratified plots, cohort comparisons, and meaningful downstream statistical analysis.
Pipeline Outputs
After the pipeline completes, CompuBio generates a rich set of outputs optimized for exploration, reporting, and integration with downstream tools.
Tabular Outputs
- Abundance tables (taxa × samples)
- Normalized count matrices
- Top taxa summaries
Visual Outputs
- Relative abundance bar plots
- Stacked bar charts across samples or groups
- Composition summaries at multiple taxonomic levels
All outputs are exportable for use in statistical testing, machine learning, or external visualization tools.
Typical Workflow
- Upload sequencing reads and optional metadata
- Select a microbial profiling pipeline
- Review quality control metrics
- Explore microbial composition plots
- Export abundance tables and visual results
You can rerun analyses with different parameters to explore how preprocessing or normalization choices affect microbial composition.
Best Practices & Notes
- Use unique, stable sample IDs across all inputs
- Apply consistent trimming and QC parameters across cohorts
- Interpret low-abundance taxa cautiously
- Always contextualize results with experimental metadata
Changing preprocessing or reference parameters between runs can make results non-comparable. Use consistent settings for cohort-level analysis.
What’s Next?
- Diversity and differential abundance analysis
- Functional profiling modules
- Integration with AI-driven classification and prediction models