Skip to main content
Support

Browse by category

All categories
Home

Soy and legumes from Denmark

Koji legume pastes outside soy (pea, lentil, fava, chickpea, lupin and cowpea misos)

Cooks and food scientists now make miso from legumes other than soy, using the same method of koji, cooked legume paste and salt left to ripen for months. Noma in Copenhagen made misos from yellow peas, Gotland lentils and fava beans, and a Portuguese study ripened chickpea, lupin and cowpea misos for 18 months.

Fully documentedSoy and legumes18 months

Recipe

Amounts from a sourcePer 100 g of the 3 % salt miso in Santos 2025. The 12 % salt version uses 65.1 g legume, 20.9 g koji, 12 g salt, 1 g miso and 1 g culture.
Cooked legume paste (chickpea, lupin or cowpea)72.9 g[santos2025]
Rice koji (Aspergillus oryzae)20.9 g[santos2025]
Sea salt3 g[santos2025]
Unpasteurised miso2.2 gBackslop from an older miso[santos2025]
Yeast and lactic bacteria culture1 g[santos2025]
Starterdefined culture: rice koji, plus backslopped miso and added yeasts and lactic bacteria

Total time across the recorded steps: 18 months

Start
Drag the bar or press Play
shapes, not to countAspergillus oryzae19–21 °C

What changes: Koji mould grows over steamed rice spread on a tray

Step 1 of 4
Make rice koji
Time
Not documented
Temperature
19–21 °C
Reported
Salt
None given
Acidity
Not documented
Which microbes the steps name, step by step
Aspergillus oryzae
Moulds
Yeasts
Lactic acid bacteria

The rice was steamed at 100 °C for 30 min, seeded at 1 × 10⁴ spores/g and turned three times a day. Output: rice koji.

Flavour
Not documented
Texture
Not documented
Colour
Not documented

Microbes at work in this step

Aspergillus oryzae grows over the steamed rice

Method

Measured measured in a studySet fixed by a specification or tested recipeReported one source reports itTypical a general value, not sourced for this fermentUnconfirmed not yet checked against a source

At a glance

Cooks and food scientists now make miso from legumes other than soy, using the same method of koji, cooked legume paste and salt left to ripen for months. Noma in Copenhagen made misos from yellow peas, Gotland lentils and fava beans, and a Portuguese study ripened chickpea, lupin and cowpea misos for 18 months. Made in 4 steps over 18 months. The longest step, pack into salted jars and ripen, takes 18 months (Santos R 2025) and runs at 19–21 °C (Santos R 2025). Salt is 3–12 % (Santos R 2025). Aspergillus oryzae, grown on rice as koji and Added yeasts and bacteria: Zygosaccharomyces rouxii, Debaryomyces hansenii, Candida versatilis, Lactiplantibacillus plantarum, Leuconostoc mesenteroides, Tetragenococcus halophilus do most of the fermenting.

Key numbers
Main step18 monthsReported[santos2025]Pack into salted jars and ripen
Temperature19–21 °CReported[santos2025]Pack into salted jars and ripen
Salt3–12 %Reported[santos2025]Mix legume, koji and salt
AciditypH 5Measured[santos2025]Pack into salted jars and ripen

What makes this one different

The method is ordinary miso; only the legume changes. Noma's work found that the legume shaped the bacteria most, while place shaped the fungi slightly more.

Related ferments

Variants

  • Noma yellow pea, lentil and fava misosMade at Noma in Copenhagen and Inua in Tokyo, some with nixtamalised legumes; also a rye bread miso.
  • Lupin misoSee lupin-ferments.
  • Grass pea sweet misoGrass pea paste, koji and sea salt in a 10:9:1 mass ratio, packed into salt-coated jars and kept at 30 °C for 4 months, then at 20 ± 1 °C [simoes2023].

Safety

Lupin seeds carry quinolizidine alkaloids, which are toxic. In the Portuguese study the lupin was soaked for 10 days, with the water changed twice a day, to reduce the alkaloid content before it was cooked (Santos 2025). The study does not report alkaloid levels in the finished miso. In a separate test of home debittering as printed on the packets, alkaloids fell 94–99 %, but all eight dried lupin samples stayed above the 200 mg/kg limit Australia and New Zealand set for lupin flour (Peloso 2026), so this is not a tested home method for bitter lupin. In that survey most brined lupins and all ready-to-use products met the limit. Grass pea holds the neurotoxic amino acid ODAP. Eating it almost alone for months can cause neurolathyrism, a spastic paralysis of the legs, though one review calls it safe within a normal diet (Singh and Rao 2013). Fava beans hold vicine and convicine, which can cause favism, an acute haemolytic anaemia, in people with G6PD deficiency (Khazaei 2019). The 3 % salt batch is below the 5–13 % salt Santos gives for traditional miso, and the paper reports no pathogen test. The FDA's fish-products guidance lists Clostridium botulinum type A and proteolytic types B and F as able to grow above pH 4.6 and up to 10 % salt in the food's water. The chickpea misos ended near pH 5 in sealed jars, so do not treat this research recipe as a tested home recipe.

History and sources

Every number is labelled with how it is known. How this atlas is made

A grass pea sweet miso made in Portugal is described in a 2023 paper, which cites Santos et al. (2021) for the recipe (Simões 2023). Kothe et al. (2024) analysed eight novel misos made at Noma in Copenhagen and six at Inua in Tokyo, from yellow peas, Gotland lentils, fava beans, rye bread and soybeans. Santos et al. (2025) ripened chickpea, lupin and cowpea misos for 18 months.

  1. [santos2025] Santos R., Parente B., Mota M., Raymundo A., Prista C. 2025, Fermentation process evaluation of a sustainable and innovative miso made from alternative legumes, Foods 14(23):4131paper
  2. [kothe2024] Kothe C.I., Carøe C., Mazel F., Zilber D., Cruz-Morales P., Mohellibi N., Evans J.D. 2024, Novel misos shape distinct microbial ecologies, Food Research International 189:114490 (abstract)paper
  3. [peloso2026] Peloso M., Prizio I., Minkoumba Sonfack G., Molgora E.B., Caprai E. 2026, Quinolizidine alkaloid profiles in lupin-based products, monitoring of the Italian retail market and efficacy of the debittering process, Foods 15(13):2269, doi:10.3390/foods15132269paper
  4. [singh2013] Singh S.S., Rao S.L.N. 2013, Lessons from neurolathyrism, a disease of the past and the future of Lathyrus sativus (Khesari dal), Indian Journal of Medical Research 138(1):32–37paper
  5. [khazaei2019] Khazaei H. et al. 2019, Eliminating vicine and convicine, the main anti-nutritional factors restricting faba bean usage, Trends in Food Science and Technology 91:549–556paper
  6. [fda-app4] US FDA 2011, Fish and Fishery Products Hazards and Controls Guidance, 4th ed., Appendix 4, Bacterial pathogen growth and inactivation, Table A-1gov
  7. [simoes2023] Simões S., Carrera Sanchez C., Santos A.J., Figueira D., Prista C., Raymundo A. 2023, Impact of grass pea sweet miso incorporation in vegan emulsions, rheological, nutritional and bioactive properties, Foods 12(7):1362, doi:10.3390/foods12071362paper