# Niche-Resolved Metagenomic Enrichment Figure · Yila

> A ternary plot places every taxon by its preference for tumour, stroma or blood, and stacked bars, boxplots, a pathway chart and three pairwise violin panels then take that split apart by function, abundance and statistics.

URL: https://yila.ai/templates/microbiome-enrichment-network-figure

The page is built around one question: how does the same microbial community differ between tumour, stroma and blood? The ternary plot gives each taxon a position, and the closer it falls to a vertex, the stronger its preference for that compartment. Stacked bars then compare how functional categories are annotated across the three, boxplots show absolute abundance once taxa are grouped by preference, the pathway chart names the functions that drive the difference, and three violin panels carry the pairwise comparisons with their sample sizes. A reader should finish knowing which compartment dominates the metabolic functions.

- **Ternary plot**: Panel a labels its three vertices as tumour, stroma and blood, and colours every point by the vertex it sits nearest.
- **Stacked annotation bars**: Panel b gives one row per functional category, with the three compartments drawn in exactly the same colours as the ternary plot.
- **Preference boxplots**: Panel c groups taxa as blood-, stroma- and tumour-enriched or balanced, and plots mean abundance along the axis.
- **Pathway enrichment chart**: Panel d ranks the pathways by enrichment from high to low, each bar coloured by the compartment that it belongs to.
- **Pairwise violins**: Panel e sets three violins side by side, each one naming the two compartments compared and marking the mean with a bar.
- **Sample size and P values**: Written under each violin in one consistent format, with the statistical test itself named in the caption.
- **Shared palette**: One fixed colour per compartment across all five panels, which lets the shared legend appear just once on the page.

- Paper mechanism figure, Graphical abstract, Academic poster, Grant proposal

- **A pathway label is overlapped**: One tick label in the enrichment chart is written over its neighbour, so the pathway behind it cannot be read at all. Overlaps like this only get worse at print size, so check every axis label before layout.
- **Violin titles collide with numbers**: The three panel titles along the bottom sit on top of the sample sizes and P values, and the middle one is barely legible at all. Raise the titles, or move the statistics into clear space inside each panel.
- **Relative or absolute enrichment**: The ternary plot is compositional while the boxplots show mean abundance, and the two can disagree. Say in the caption which claims rest on composition, so that nobody reads them as absolute change in load.
- **Contamination controls for blood**: Low-biomass blood metagenomes readily pick up reagent and environmental taxa, so reviewers will ask for negative controls and for the decontamination steps. Both belong in the methods and in this figure's caption.

## FAQ

### Can this go straight into a paper?

Use it as a reference for multi-panel layout rather than as data. The points, bars and statistics are AI-generated illustration, so redraw every panel from your own metagenomic results, and fix the overlapping pathway label and the violin titles long before the figure goes anywhere near a submission.

### Can elements in the figure be changed?

Not element by element, but you can keep refining it by chatting in Yila: rename the compartments to your own sampling sites, drop the pathway chart, replace the violins with paired scatter plots, or keep only the first two panels, then generate a new version from that description.

### What format is delivered?

An AI-generated PNG image in a 4:3 landscape layout. It is not a vector graphic, and with five panels on one page the pathway names and the axis ticks blur once the figure is reduced, so always check it at the column width of your target journal first.

### Is it free?

Viewing this page and the sample figure costs nothing. Generating a version in Yila consumes credits, and a new account receives 2,500 credits, which is enough to try several panel combinations and then refine whichever arrangement supports your own conclusion best.

### I only have two compartments, not three. What then?

Tell Yila how many groups you sampled and ask for the ternary plot to become a two-axis scatter or a volcano plot. The other four panels stay as they are with a two-colour palette, and the order of evidence down the page still holds together.

### Can I keep only the ternary plot?

Ask for the ternary plot and its vertex labels alone, without the other four panels around it. On its own it suits a defense slide, where a single clear claim about niche preference lands better than function, abundance and statistics all presented at once.

---

- [Niche-Resolved Metagenomic Enrichment Figure](https://yila.ai/templates/microbiome-enrichment-network-figure)
- [llms.txt](https://yila.ai/llms.txt)
