lab / capsicum-lab

Capsicum Lab — Chilli Pepper Data Explorer

An early-stage dataset and exploration tool for the diversity of Capsicum. It combines two real sources: species-level taxonomy from the GBIF Backbone Taxonomy, and 5,270 genebank accessions with passport and fruit-morphology data from the WorldVeg Capsicum characterization dataset on Genesys.

This is an MVP: genebank trait values come from a single collection rather than the full species range, several measurement units are inferred rather than confirmed against a descriptor legend, and heat data uses reported Scoville ranges from editorial sources (flagged as low confidence accordingly). Missing values are left blank rather than guessed.

Heat landscape

Reported Scoville (SHU) ranges for 0 cultivars across 0 Capsicum species, on a logarithmic scale — chilli heat spans several orders of magnitude, so a linear axis would crush everything below the superhots into a single line. These are editorial/reported ranges (sourced from Wikipedia and common references), not lab-measured capsaicinoids; treat them as indicative rather than precise.

Loading heat data...

Heat by fruit shape

The same cultivars, grouped by their editorial shape_category and plotted by a log-scale midpoint of each SHU range (fairer than a plain average for ranges spanning orders of magnitude). Gray boxes show the quartiles/median per shape; dots are individual cultivars, colored by species — the same colors as the chart above.

How much should you trust this?

A compact summary of the two charts above, rather than a claim of statistical rigor.

  • 0 cultivars have a numeric SHU range, out of 0 total — sourced from Wikipedia's List of Capsicum cultivars and a small project pilot, not lab-measured capsaicinoids. low confidence
  • Wider reported ranges tend to come with higher reported heat (Spearman r = n/a) — superhot SHU figures anywhere on this page are order-of-magnitude estimates, not precise measurements.
  • Some shape categories have very few cultivars (marked above) — their medians can shift a lot with one more or fewer data point.
Explore the wider dataset9 more views: size, sweetness, correlations, origin, and the full browsable datasetClick to expand ▾

Shape and size explorer

Fruit length vs. fruit width for every accession with both measurements recorded, colored by species. Click a legend entry to isolate a species.

Loading morphology data...

Fruit weight distribution

Fruit weight for every accession that has it recorded, on a logarithmic x-axis and stacked by species.

Size vs. sweetness

Fruit width vs. soluble solids (°Brix), faceted by the four best-sampled species. If bigger fruit were just as sweet, these clouds would be flat; a downward slope means bigger, more water-diluted fruit measure less sugar per volume.

Morphometric correlation heatmap

Pearson correlation across the 9 real continuous accession traits. seed_countis deliberately excluded — it's an ordinal 1-3-5-7-9 genebank characterization score, not a measured quantity, so a correlation on it would be invalid.

Loading correlation data...

Species trait fingerprint

Median trait values per species (accessions only, species with 20+ records) — a domestication signature: early flowering and large fruit in the cultivated species vs. the opposite in wild relatives.

Loading species fingerprint data...

Occurrence map

Where Capsicum has been observed or collected, from GBIF occurrence records (up to 25 georeferenced points per species). These are specimen and observation locations — including cultivated plants far outside the genus's native range — not native-range or cultivar-origin claims. The busiest species are colored individually; the rest are grouped as “Other species”.

Loading occurrence data...

0 georeferenced occurrences. Occurrence and media licences vary per record (see GBIF).

Origin of the collection

Genebank accessions by raw MCPD country-of-collection code (origins.csv), tile size = accession count, tile colour = dominant species. This is a different data source than the occurrence map above: this is where the genebank got its seed samples, not where Capsicum has been biologically observed — no coordinates exist for this table, only country codes.

Loading origin data...

Data coverage

What proportion of the dataset comes from each source, how complete the key fields are for genebank accessions, and what major data categories are still missing.

Loading coverage data…

Browse the dataset

Loading dataset size... Search by name or filter by entity type; select a row to see everything recorded for it, including sources.

0 matching
What do these types mean?
Species
An accepted Capsicum species from the GBIF Backbone Taxonomy (e.g. C. annuum). Taxonomic record only — no fruit measurements.
Accession
A physical seed sample held by a genebank (WorldVeg/AVRDC, Taiwan). Has origin country and fruit measurements. One species can have thousands of accessions.
Cultivar
A named, human-selected variety (e.g. Jalapeño, Habanero, Scotch Bonnet). This is where heat (SHU) data lives.
Generic category
A broad food grouping from USDA FoodData Central (e.g. “Peppers, hot chili, raw”). Used for nutritional reference data, not a specific variety.
Page 1 of 1