The Finnish Disease Heritage: 39 rare diseases, one bottleneck, and what a DNA file can show

Thirty-nine diseases and four tests

A disease does not join the heritage because it is rare or because a Finnish family has it. The criteria go back to the geneticist Reijo Norio, and the 2022 review restates them as four tests^[1].

1. It is more common in Finland than in the rest of the world together, or markedly over-represented in Finland.
2. It has been diagnosed in more than ten core families.
3. It has a Finnish founder mutation.
4. That mutation's frequency differs significantly between Finns and other populations.

The count depends on when you look. The Finnish Disease Database, FinDis, which began in 2004 with 30 diseases^[3], still introduces the heritage as 36^[2]. The 2022 review counts 39, because three more met the criteria after 2003: spinal muscular atrophy of the Jokela type, a motor neuron disease of adults; frontotemporal pachygyria, an intellectual disability; and cytosolic PEPCK deficiency, a treatable metabolic disorder^[1].

Of the 39, 34 are recessive: they appear only in a child who inherits a disease-causing copy from each parent, and the parents, with one copy each, are healthy carriers. Three are dominant and two X-linked^[1]. The recessive ones are why the heritage matters for family planning.

How a population gets its own diseases

For most of its history Finland had few people. In the twelfth century the population was about 50,000. It reached 400,000 by the middle of the seventeenth century, and the great famine of the 1690s then killed about a third of it^[2].

How people settled mattered as much. From the sixteenth century, groups of farmers from a small area of South Savo moved into the centre of the country, then the west and finally the north, clearing land by fire. Within a century the inhabited area doubled, and many north-eastern settlements grew from 40 to 60 founding families, with no further immigration, until the Second World War^[2]. Norio's standard review of the heritage gives this internal migration a section of its own^[5].

A founder effect is chance, not selection. If one of a few dozen founding families carried a recessive variant, its descendants could inherit it, and because carriers are healthy, nothing removed it as the settlement grew. Where everyone descends from the same few families, two carriers of the same variant meet far more often. The 2022 review puts the enrichment down to bottlenecks too recent for selection to have worn these variants down^[1].

The history is well recorded. Finnish parish records of births, deaths, marriages and moves go back to 1640^[1]. When Varilo and colleagues traced 18 Finnish families with CLN5 disease through church records, the pedigrees merged again and again, and they dated its single founder mutation to 20 to 30 generations ago, about 500 years^[6], close to when the internal migration began.

One variant per disease

Because each disease goes back to one or a few ancestors, Finnish patients mostly carry the same variant, inside a long stretch of shared DNA whose length reflects the variant's age^[2]. Elsewhere the same diseases usually have other causes: in non-Finnish patients the causative variant typically differs from the Finnish one^[1].

A single main variant makes a disease easier to test for: a test aimed at that single position finds most Finnish carriers, and so does any sequencing file that covers it. For a person of Finnish ancestry, not carrying the founder variant removes most of the chance of being a carrier, and for some diseases nearly all of it^[2].

What the founders did not bring

The founder effect works in both directions. Variants the founders did not carry stayed out, or arrived later and stayed rare. Phenylketonuria, galactosemia and cystic fibrosis, recessive diseases well known elsewhere in Europe, are exceptionally rare or even absent in Finns^[1]; as HUS, Helsinki University Hospital, puts it, some diseases very rare elsewhere are more common in Finland, while others are significantly rarer^[4]. See our pages on phenylketonuria and cystic fibrosis.

Where in Finland

Geneticists map the diseases by the birthplaces of patients' grandparents, which still show the old clustering after decades of moving^[2]. For most recessive heritage diseases those birthplaces cluster in the late-settlement region, the inland north and east settled from the 1500s. A few of the commoner ones, among them cartilage-hair hypoplasia, AGU and congenital nephrosis, occur throughout the country; Meckel syndrome and diastrophic dysplasia follow the older settlement of the west. Two are strictly local: Northern epilepsy in Kainuu, near the eastern border, and CLN5 disease in Southern Ostrobothnia^[2].

Genetic data have since measured what the birthplaces suggested. Comparing 43,254 Finns with 16,060 people from neighbouring countries, Martin and colleagues found that two unrelated Finns share, on average, at least one long stretch of DNA from a common ancestor; north-eastern Finns share several times more than south-western ones, and two people carrying the same rare disease-causing variant share about ten times more than other pairs from their area^[10].

Beyond the classic list: 104 genes

The heritage is a list of diseases chosen for being Finland-specific, not a list of the recessive variants Finns most often carry. Going through gnomAD v4, whose 800,000 people include about 30,000 Finns, researchers at HUS and the University of Helsinki found 104 genes with a cumulative Finnish carrier frequency above 0.5 percent^[4].

Our panel also reads 26 heritage genes, a coincidence: HUS counted heritage genes above a frequency line, our panel those a sequencing file calls reliably, and the lists need not match. The paper behind the announcement counted 324 genes with a carrier frequency of at least 1 in 200 in some ancestry group, the line at which the American College of Medical Genetics and Genomics recommends offering screening for severe or moderate conditions^[12]. An earlier analysis found 18 heritage genes at 0.5 percent or more in Finns missing from the ACMG list^[8].

Some variants commoner in Finland than anywhere else never made the list, because the disease they cause is not concentrated here. The review's example is LCHAD deficiency, a metabolic disorder: Finnish carrier frequencies for one of its variants run from 1 in 132 in rural to 1 in 365 in urban areas, but the disease occurs worldwide, so it fails the first test^[1]. And new founder variants are still turning up: in 2026 a FinnGen study found a recessive variant in XXYLT1 behind an inherited retinal dystrophy^[13], which our report reads from the study; our XXYLT1 page has the numbers.

The pattern reaches beyond rare disease. FinnGen, which aims to study 500,000 Finns, reported in 2023 that of 148 coding variants its analysis tied to disease, 91 were uncommon in other Europeans and 62 of those were more than twice as common in Finland^[14].

Which tests can read the heritage

A genotyping chip, the technology behind 23andMe and AncestryDNA, measures a fixed list of positions chosen in advance; a whole-genome file covers the genes themselves. For diseases that come down to one variant, either can work in principle.

23andMe describes its own negative result carefully: the person does not have the variants it tested, and could still have one the test does not cover^[15]. We do not issue carrier results from chip files at all, because chip calls grow less reliable the rarer a variant is: against sequencing in the UK Biobank, only 16 percent of chip calls for variants rarer than 1 in 100,000 were confirmed^[16], and most heritage founder variants are rare outside Finland.

The 39 diseases and what a whole-genome file reads

Our carrier panel matches 26 heritage genes against ClinVar's pathogenic and likely pathogenic entries, so a carrier of a rarer variant in the same gene is found too. Each is there because its founder variant is a short change that short-read sequencing calls reliably; the MKS1 one, a 29-letter deletion inside an intron, is borderline, and its card says so.

These frequencies are rough, and the national and regional figures for one disease can differ several-fold, as Northern epilepsy's do.

The file type changes how much a quiet result means. A gVCF records where the reads matched the reference, so a card can say a gene was examined and nothing found; a plain VCF's silence cannot be told apart from a gene never read, as our guide to genome files shows on real files.

Couple screening in Finland today

Predictive carrier screening for couples is not currently offered in Finland. HUS said so in 2024, announcing its estimate that about 3 percent of couples with Finnish ancestry are carrier couples for a recessive or X-linked rare disease, both partners carrying a variant in the same gene. Its researchers wrote that the results allow the question to be considered for Finland, and noted a limit no screening removes: some diseases come from new variants that neither parent carries^[4].

Our partner view sets two people's carrier results side by side, each from their own whole-genome file, and says where both carry a variant in the same gene. It is not clinical carrier screening. It reads a fixed panel and the variants ClinVar lists, so a gene where neither file shows a variant is less likely to be shared, not ruled out, and a result from it is not a basis for a reproductive decision on its own.