Quickstart

From the full catalogue to an exported shortlist

This walkthrough takes you through one complete pass of the discovery platform: build a query, run it, triage the matches, and export a shortlist. It assumes you have read Key Concepts (in particular what an iBGC is).

When you first open the platform you’ll see a short welcome and the option to take an interactive tour. The tour mirrors the steps below; you can replay it any time.

1. Build a query

The top strip holds the filter chips. Add chips to narrow the catalogue. For a first pass, try a single restriction:

  • Click the Taxonomy filter and pick a phylum or genus of interest (for example Actinomycetota).
  • Optionally add a Biome chip (for example Marine) or a BGC class chip.

The badges above the chips show the size of the full catalogue — validated BGCs, integrated BGCs, predicted BGCs, genomes, and metagenomes. Nothing is filtered yet.

2. Run the query

Press Run Query. A banner reports how many iBGCs match your filters. Until you run, the result panels stay empty — this is deliberate, so you set up the whole query before anything loads.

3. Triage in the BGC Roster

The left panel shows results three ways via tabs. Start on the BGC roster — a sortable table, one row per iBGC.

  • Click the Novelty column header to sort. The most novel candidates (least like any validated cluster) rise to the top.
  • Each row shows the iBGC accession, predicted class, size, novelty, domain novelty, the detection tools, and the parent assembly. Badges flag Validated, Type Strain, and Partial iBGCs.

Switch to the Variables map tab to plot two metrics against each other (e.g. Novelty vs Size), or UMAP to see how the matches group — similar iBGCs sit close together, coloured by gene cluster family.

4. Inspect and compare

  • Left-click an interesting row (or a point on a map). It loads into the Compare panel on the right, showing its location, source predictions, GCF, predicted compounds, and scores, with a gene-level region plot.
  • Right-click a row for more actions: Set as reference iBGC pins it to the top panel so you can compare others against it; Find similar iBGCs expands your search around it; Add to shortlist collects it.
  • Click any gene in a detail panel to load its protein — sequence and domain annotations — into the Protein Information panel at the bottom.

5. Shortlist the best candidates

Right-click promising iBGCs and choose Add to shortlist, or use Add all to shortlist in the roster footer to add everything matching the current query (up to the shortlist cap). The shortlist lives in your browser and persists between sessions.

6. Generate a report and export

Open the shortlist and click Generate Report. The platform materialises your selection into a standalone report page with summary charts (domain composition, GCF distribution, biome and taxonomy breakdowns, score distributions) and tables. From there, download:

  • GenBank (.gbk) files — one per source BGC, grouped by iBGC, ready for annotation tools.
  • An iBGC table (TSV) and a tidy JSON bundle of the underlying data.

That’s a full pass. From here you can refine filters and re-run, switch to a targeted search, or load your own assembly to compare it against the catalogue.