Zoh Science Illustrated · By Zoh Probiotics · Evidence checked 7 October 2026
The tiny version
A name can sound alarming or reassuring before we know what it means. With yeasts, read the full name and the context first. Finding an organism in a study is not the same as diagnosing an infection, approving a starter or proving a health benefit.
What do Candida and Yarrowia mean in a kefir paper?
They are yeast names reported in identification studies. In the Greek kefir survey by Kalamaki and Angelidis, both genera appeared among the organisms identified. That is a description of study samples, not a statement about every kefir. [1]
These names deserve more context than a label such as “good yeast” or “bad yeast”. Read the species, sample and method before interpreting the finding. A food survey cannot diagnose a person, and the fact that an organism came from food cannot settle every safety question.
Name
Which species or strain was identified?
Setting
Was the sample a food, isolate or clinical specimen?
Question
Was the study testing identity, safety or an effect?
Meet the microbe · a shape guide
Yarrowia lipolytica
One organism, different growth forms
ATCC culture 20460 is described with rounded, oval and elongated cells. It can also form pseudohyphae and true hyphae. Here we show only individual yeast-like cells, not its full range of growth forms.
This is a Yarrowia example, not a universal Candida portrait. Appearance varies with growth conditions.
ATCC 20460: morphology description. Conceptual illustration, not microscopy. Colour and surface texture are artistic; not to scale. Shape cannot identify a strain, demonstrate health effects or verify Zoh product composition.
What has been observed about Yarrowia?
Kazou and colleagues reported Yarrowia lipolytica among yeast isolates from their kefir investigation. The study used culture-based and molecular approaches to examine homemade and industrial samples. [2]
The presence of an identified isolate helps describe that sample. It does not establish that the organism is needed in all kefir, nor that adding it separately improves the drink. A community survey is not a formulation guide.
This also explains why the word “contains” needs care. A statement about a surveyed food and a statement about an actual product sold today require different evidence. None of the papers here certifies Zoh's culture composition.
Why can one yeast have several names?
Taxonomic names can reflect different stages in the history of classification. Douglass and colleagues used genome comparisons to show that organisms described as Candida krusei and Pichia kudriavzevii belong to the same species. The paper also discusses the name Issatchenkia orientalis. [3]
The practical consequence is simple: changing the name on the page does not create a new organism. It can also hide relevant information if readers search only one of the names. A carefully edited microbial list checks synonyms before counting entries.
The clinical lens
What can an identification study establish?
It can report the organisms identified under its methods and sampling conditions. It does not automatically establish the consequences of eating a particular strain, the reason for a person's symptoms or the suitability of intentional use in another food.
Safety and benefit are also separate assessments. EFSA's QPS framework considers taxonomic identity, knowledge and safety concerns; it does not substitute for every product-level assessment. [4] A health benefit requires evidence for the intended intervention and outcome, not just a reassuring category.
What should not be inferred from these papers?
They do not demonstrate that kefir treats a Candida infection. They do not allow a photograph of a jar to identify a species. And they do not show that every organism listed in a survey is an appropriate starter ingredient.
The reverse shortcut is unhelpful too: encountering a familiar genus in a food paper is not a diagnosis. The clinical and food contexts must be assessed on their own evidence.
What it means for your health
Use this information to ask more precise questions, rather than to self-diagnose or select a treatment. A suspected infection needs clinical assessment. Choosing a ferment from a microbial name list is not a substitute.
For food decisions, choose a traceable starter and use the intended method. Do not buy an isolated yeast simply because it appeared in a kefir paper. This article explains research terminology, not a home method for selecting organisms.
What to do with an unexpected batch
If there is unexpected surface growth, stop using the batch and seek guidance. Record the starter, milk, temperature, timings and storage history. A photograph can help explain the problem to a supplier, but it cannot establish the organism's identity or clear a batch as safe.
Keep the question practical: “Does this result fit the recommended process, and what should I do with this batch?” is more useful than trying to decide which yeast it resembles from a stylised picture.
Next: avoid counting the same organism twice
The naming issue continues with Debaryomyces and Issatchenkia, including a simple way to audit a long species list.
Questions people ask
Does kefir treat a Candida infection?
This article provides no evidence for that claim. An infection needs appropriate clinical care.
Can I tell which yeast is present by looking at the jar?
No. Identification requires suitable laboratory methods.
Research notes: follow the claim
Selected sources for this educational explainer, not a systematic review. Study types are labelled; no independent clinical review is claimed.
- Yeast identification study. Kalamaki MS, Angelidis AS. Isolation and molecular identification of yeasts in Greek kefir. International Journal of Dairy Technology. 2017;70:261–268. Source and bibliographic record.
- Culture and sequencing study. Kazou M et al. Zooming Into the Microbiota of Home-Made and Industrial Kefir Produced in Greece Using Classical Microbiological and Amplicon-Based Metagenomics Analyses. Frontiers in Microbiology. 2021;12:621069. Source and bibliographic record.
- Comparative genomics. Douglass AP et al. Population genomics shows no distinction between pathogenic Candida krusei and environmental Pichia kudriavzevii: One species, four names. PLoS Pathogens. 2018;14:e1007138. Source and bibliographic record.
- Safety assessment framework. European Food Safety Authority. Qualified presumption of safety (QPS). Current framework consulted 7 October 2026. Source and bibliographic record.
All topics · Recipes · Science corner
Keep the story going
