Interactive Report Demo¶
A live, fully interactive TaxTriage report built from the example dataset - the same artifact the pipeline writes to multiqc/heatmap.html at the end of a run. Nothing is uploaded and nothing is computed server-side; the whole report is a single self-contained HTML file carrying its own data.
Open in a new tab Read the report guide
What you're looking at¶
| Tab | Shows |
|---|---|
| Summary | Per-sample overview with the top-scoring organisms and QC flags |
| Heatmap | Organism × sample matrix; switch the cell value between TASS, abundance, coverage and depth |
| TASS | Score decomposition - how each component contributed to the final confidence |
| Coverage | Per-reference coverage tracks with the conflict regions marked |
| Sunburst | Taxonomic composition, drillable from kingdom down to species |
| Proteins / VF-AMR | Virulence factor and antimicrobial resistance gene hits |
| Explore | Free-form filtering across every metric in the run |
Full documentation of each panel lives in Interactive Report, and the scoring behind the heatmap values is described in TASS Scoring.
Try the sample QC rules¶
The demo opens with two sample QC rules already active, so the feature is visible rather than hidden behind a dialog. They flag three of the six samples - one for shallow read depth, three for having fewer than four distinct organisms above TASS 99 - and you can see the effect as an outlined column in the Heatmap, a tinted group header in the Table and Summary, and a badge in Metadata & Mapping. Hover any marker for the exact reason.
Open Sample QC / Flags → Filters & flags… in the right-hand panel to edit them: change a threshold and watch the flagged list update as you type, add a rule on any metadata column or detection metric, or switch a rule from flag it to flag & hide it to drop those samples from every view. Remove all rules clears them entirely.
These thresholds are illustrative
They are tuned to trip on the example dataset so the markers are visible. In a real run the defaults come from the pipeline's --report_flag_* parameters - see Report Sample-QC Flags.
This is example data
The demo is built from the pipeline's test dataset so the layout and interactions are representative, but the organisms and scores are not from a real clinical sample. Run the pipeline on your own data to generate a report like this - see Quick Start.