Learn Read your report
Carrier status vs your own risk: what one copy of a recessive variant means
A carrier result reads like a warning about you. For most recessive conditions it is a fact about a gene you could pass on: one working copy is normally enough. There are known exceptions, and the report labels the results that are not plain carrier status differently.
Key takeaways
- One disease-causing copy of a recessive gene usually leaves the carrier unaffected, and carrying some is ordinary: one screen found 2.8 per person on average[2].
- Some carrier states have measurable effects on the carrier: sickle cell trait, cystic fibrosis carriers, GBA1 carriers and Parkinson's disease, the Finnish TYROBP deletion and Alzheimer's disease. Absolute risks are mostly small[6, 3, 8, 11].
- How common a carrier variant is depends on ancestry. Two Finnish founder variants each reach about 0.4 to 0.7 percent of gene copies in Finnish genomes, while the commonest cystic fibrosis variant is almost ten times rarer in Finns than in other Europeans[15].
- A chip cannot give carrier status reliably: in UK Biobank only 16 percent of chip calls for very rare variants were confirmed by sequencing[17]. The report reads carrier status from whole-genome files only.
You have two copies of almost every gene, one from each parent. A recessive condition appears when both copies are altered, and the parents of an affected child each carry one altered copy but typically show no signs of the condition[1]. Carrying one is ordinary: a sequencing screen for 448 severe recessive childhood diseases found an average of 2.8 disease-causing variants per person, from none to seven[2].
One working copy is usually enough
A recessive variant stops one copy of a gene from doing its job. The other copy still works, and for most genes the protein that one copy makes is enough. Cystic fibrosis is the textbook case: carriers have about half the CFTR channel activity of non-carriers[3], and they do not have cystic fibrosis. That is the whole meaning of a carrier: a person with one disease-causing copy of a recessive gene and one working copy.
Three terms this guide relies on
- Recessive
- A condition that appears when both copies of a gene are altered. One altered copy alone typically causes no signs[1].
- Heterozygous
- Carrying a variant on one of your two copies. In a VCF the genotype is written
0/1. Homozygous means the same variant on both copies,1/1. - Phase
- Which copy each variant sits on. A standard genome file lists two variants in a gene without saying whether they share a copy; two on different copies count as two altered copies.
The carrier result matters for the next generation. When both parents carry a variant in the same recessive gene, each pregnancy has a 25 percent chance of a child with the condition, a 50 percent chance of a child who is a carrier like the parents, and a 25 percent chance of a child with neither[4].
2.8
average number of variants for 448 severe childhood recessive diseases per person, ranging from 0 to 7[2]
52.7%
of Finns carry at least one variant in the ACMG-recommended recessive genes plus the Finnish heritage genes[5]
1 in 4
chance per pregnancy of a child with the condition when both parents carry a variant in the same recessive gene[4]
Where one copy does have an effect
Half the working protein is enough to prevent the recessive disease. It is not always enough for everything else the gene does, and large studies have found effects in carriers of several well-known recessive variants. They usually differ from the disease itself and are small for any one person.
| Carrier state | What studies found in carriers | How large |
|---|---|---|
| Sickle cell trait (one HbS copy in HBB) | More exertional rhabdomyolysis among 47,944 Black US Army soldiers, and no difference in death[6]. More chronic kidney disease in 15,975 African Americans[7]. | Rhabdomyolysis hazard ratio 1.54; kidney disease odds ratio 1.57, absolute difference 7.6 percentage points |
| Cystic fibrosis carrier (one CFTR variant) | Higher rates of 57 of 59 cystic fibrosis-related diagnoses among 19,802 carriers in US insurance claims, including chronic pancreatitis, male infertility and bronchiectasis[3]. | Chronic pancreatitis odds ratio 6.76, but 0.429 per 100 carriers |
| Gaucher disease carrier (one GBA1 variant) | Parkinson's disease: GBA1 variants were found far more often in 5,691 patients than in 4,898 controls[8]. | Odds ratio 5.43. Cumulative risk by age 80 to 85: 29.7 percent in a familial study[9], 10.9 percent among parents of Gaucher patients[10] |
| Nasu-Hakola disease carrier (one copy of the Finnish TYROBP deletion) | Alzheimer's disease and dementia among 2,231 carriers in FinnGen's 520,210 Finns, diagnosed more often and earlier[11]. | Alzheimer's hazard ratio 2.59 after adjusting for APOE e4; no absolute risk reported |
| G6PD deficiency, one copy in a person with two X chromosomes | Red cells can be mostly deficient while a standard screening test reads normal; in one malaria trial 9 of 17 such women on a higher primaquine dose had clinically significant haemolysis[12]. | Depends on which X is switched off in each red cell |
Source: The studies cited in each row[6, 7, 3, 8, 9, 10, 11, 12]. Hazard and odds ratios compare carriers with non-carriers in the same study.
The numbers need reading with care. An odds ratio of 6.76 for chronic pancreatitis sounds alarming, and the same study found the condition in fewer than one carrier in two hundred; its authors write that the vast majority of cystic fibrosis carriers will never develop pancreatitis[3]. The two GBA1 estimates differ almost threefold, and so do the designs: one recruited families that already had Parkinson's disease[9], the other asked about the parents of Gaucher patients, who were found through their children[10]. The TYROBP study found its carriers through a marker inherited with the deletion, not the deletion itself[11].
The X chromosome is the other exception. A person with one X has only one copy of each X-linked gene, so one disease-causing variant is enough to cause an X-linked recessive condition[1]. A person with two X chromosomes and one such variant typically has no or very mild signs[4], but each cell switches off one X at random, and the mix varies. That is how a G6PD carrier can have mostly deficient red cells[12].
Why carrier frequency depends on ancestry
A recessive variant rarely harms the carriers who pass it on, so selection barely touches it while it stays rare. Its frequency then follows a population's history. In a small founding group, whatever variants the founders happened to carry can become common among their descendants, and others are lost. The Finnish Disease Heritage is the best-studied example: diseases caused by single major variants, usually recessive, enriched by several past episodes of genetic drift[13]. Our guide to the Finnish Disease Heritage tells that history in full, and the runs of homozygosity guide shows the same history inside single genomes.
Selection can push the other way. One copy of the sickle variant protects against malaria: in a meta-analysis, people with sickle cell trait had an odds ratio of 0.09 for severe falciparum malaria compared with people without it[14]. Where malaria was common, carriers survived it more often, and the variant became common too.
| Variant | Finnish | Non-Finnish European | African / African American | Ashkenazi Jewish | East Asian |
|---|---|---|---|---|---|
| CFTR F508del | 0.151% | 1.43% | 0.26% | 0.403% | not seen |
| HBB sickle variant (HbS) | not seen | 0.0162% | 4.36% | not seen | not seen |
| HEXA 1278insTATC | 0.00942% | 0.0294% | 0.00483% | 1.27% | not seen |
| GBA1 N370S | 0.104% | 0.182% | 0.0193% | 2.91% | not seen |
| AGA AGU-Fin major | 0.707% | 0.00735% | not seen | not seen | not seen |
| SLC17A5 Salla (R39C) | 0.443% | 0.0294% | 0.00241% | not seen | not seen |
Source: gnomAD v4, genome callset of 76,215 people, read on 10 October 2026[15]. The Finnish group is about 5,300 genomes and the Ashkenazi Jewish about 1,700, so a frequency of one or two copies there is rough.
Each variant has a home. The two Finnish founder variants, behind aspartylglucosaminuria and Salla disease, are 96 and 15 times as frequent in Finnish genomes as in other European ones. The commonest cystic fibrosis variant runs the other way: almost ten times rarer in Finns than in other Europeans, which fits cystic fibrosis being exceptionally rare in Finland[13]. Across many genes the totals even out: the 2024 gnomAD analysis behind the 52.7 percent figure for Finns found 48.9 percent for other Europeans, and couples sharing a carrier gene at 1.4 and 0.93 percent[5]. Over more genes, HUS put Finnish carrier couples at about 3 percent[16].
What the carrier section of the report shows
The carrier panel reads 60 genes: 59 autosomal recessive and one X-linked, G6PD. 26 of them are Finnish Disease Heritage genes. In each it looks for variants ClinVar classifies as pathogenic or likely pathogenic, at any position in the gene. Findings with at least two-star ClinVar review are the main results; one-star, single-submitter matches sit in a separate, collapsed list as lower-confidence leads, and a zero-star entry is never reported. A short list of founder variants that a peer-reviewed study has shown to cause disease, but ClinVar has not yet classified, is reported at their exact positions only and marked as published research. The confidence tiers guide explains the stars.
The band on each card comes from the genotype in the file.
| Band | When it appears | What the card does |
|---|---|---|
| Carrier | One copy of a classified variant in a recessive gene | Restates the condition through a carrier lens and gives a carrier frequency |
| Two copies detected | The same variant on both copies | Says plainly that this is not carrier status and may mean the condition is present; the carrier narrative is left out |
| Two changes, phase unknown | Two different two-star variants in one recessive gene | Says the file cannot tell whether they share a copy; the two variants are never shown as two separate carrier results |
| Single X copy | A G6PD variant called on the only X | Explains that one copy on the only X usually means the deficiency itself |
Source: The carrier module and card templates the report runs on 10 October 2026.
Finding nothing has limits, and the report states them. Ten conditions are listed as gaps. Most cannot be called from a variant file, among them spinal muscular atrophy, the alpha-thalassemia deletions and fragile X, which need copy-number calling or repeat sizing; one, a Finnish FSHR variant, has not been added to the panel yet. Two of the ten, the Finnish deletions behind Nasu-Hakola disease and juvenile CLN3 disease, are read from aligned reads on GRCh38. Four genes where a clear result is less certain than usual carry a note saying so, among them GBA1, which sits beside a near-identical pseudogene, and MKS1, whose Finnish variant is a 29-letter deletion.
Below the curated panel, a wider screen looks for 98,322 specific variants that ClinVar lists as disease-causing in 1,363 further recessive and X-linked genes, in ClinVar's and ClinGen's own words. A variant ClinVar does not list yet is not looked for.
The report's labels follow the evidence, so the exceptions above are not erased. The TYROBP deletion card does not call its carrier unaffected and keeps the Alzheimer's finding behind a reveal the reader opens, and the G6PD card says what one copy means on a single X.
What each file can show
| File | Carrier panel, 60 genes | Wider screen, 1,363 genes | Partner view |
|---|---|---|---|
| Chip export (23andMe, AncestryDNA, MyHeritage) | Not run | Not run | Cannot take part |
| Plain VCF | Runs; a gene with no finding cannot be shown to have been read | Runs; says it cannot show whether a listed position was read | Takes part; its silence reads "No variant found; not examined" |
| gVCF | Runs, with a table of how well each panel gene was covered | Runs, with the share of listed positions it covered | Takes part; a covered gene can read "Examined, none found" |
| BAM or CRAM, Deep Read | Adds the TYROBP and CLN3 deletions on GRCh38 | No change | Read depth can show a gene was covered, as a gVCF can |
Source: The report modules, partner view and Deep Read as they run on 10 October 2026. Why a plain VCF's silence cannot count as a clear result is in our guide to genome files.
The partner view sets two people's carrier results side by side, each read from that person's own whole-genome file. Only two-star findings join it. Where both carry a variant in the same autosomal recessive gene and each has one, it states the one-in-four pattern once, in words; where one of them has two variants, it gives no fraction, because the one-in-four figure is for two single carriers. An X-linked gene gets neither, because the pattern depends on which partner carries it, and the view infers no one's sex. It is not clinical carrier screening, and in Finland predictive carrier screening is not currently offered to couples[16].
Why a chip cannot give carrier status
A genotyping chip measures a fixed list of positions, mostly common ones. Carrier variants are rare outside their home populations, and rare positions are where chips fail. Comparing chip calls with sequencing for 49,908 UK Biobank participants, Weedon and colleagues found that for variants rarer than 1 in 100,000, only 16 percent of 4,757 heterozygous chip calls were confirmed; 20 of 21 consumer chip files they examined held at least one false rare disease-causing variant[17].
A vendor's own carrier report is a different thing from re-reading the raw file: it covers a short list of variants the vendor has validated, such as a single variant in the SACS gene for ARSACS on 23andMe's European list, and it describes its negative result with care[18].
You do not have the variant(s) we tested. There is still a chance that you could have a variant not covered by this test.
We issue no carrier results from chip files, and a chip report says so. A rare disease-causing variant in a chip export is often a genotyping error, as our guide to false positives in raw data shows on real files.
What Aimosti would (and wouldn't) show you
On a whole-genome VCF or gVCF, the carrier panel reads 60 genes and a wider screen ClinVar's listed variants in 1,363 more. Each card gives the variant, your genotype, ClinVar's classification and stars, and a band: Carrier, Two copies detected, Two changes, phase unknown, or Single X copy. A chip export gets no carrier results. The sample report shows the whole section.
What we won't claim
We won't turn a one-copy carrier result into a diagnosis or a risk figure for you, and we won't say a carrier is unaffected where research has found otherwise. We restate ClinVar's classification and never assign our own. We won't issue carrier results from a chip file, call a gene clear on a plain VCF's silence, or claim carrier status for conditions a variant file cannot see, such as spinal muscular atrophy, the alpha-thalassemia deletions and fragile X; those are listed as gaps.
Bottom line. For most recessive conditions a carrier result describes a gene you could pass on, not your own health. A few carrier states have measurable effects, and they are known by name. Two copies, two changes and a single X copy get their own labels, and a chip gets no carrier result at all.
Questions people ask
Does being a carrier affect my own health?
For most recessive conditions, no: one working copy is enough, and the parents of a child with a recessive condition typically show no signs of it[1]. A few carrier states have measured effects, including sickle cell trait, cystic fibrosis carriers and GBA1 carriers, mostly as raised relative risks of other conditions[6, 3, 8].
Why does my report say Two copies detected instead of Carrier?
Because the file shows the same disease-causing variant on both copies of the gene. That is not carrier status, and for a recessive condition it can mean the condition is present rather than carried; the card says so and leaves out the carrier text. Whether it is present is a clinical question a genome file cannot settle.
Can I get carrier status from my 23andMe or AncestryDNA raw data?
Not reliably. Chip calls of very rare variants were confirmed by sequencing only 16 percent of the time in UK Biobank[17], and a chip measures only the positions it was built for. A vendor's own carrier report covers a short validated list. Our report does not issue carrier results from chip files.
Why is a carrier frequency different on a Finnish site than on an American one?
Because carrier frequencies belong to populations. The AGU-Fin major variant is on 0.707 percent of gene copies in Finnish genomes and 0.00735 percent in other European ones, while the commonest cystic fibrosis variant is about ten times rarer in Finns[15].
Is the partner view carrier screening?
No. It compares two people's results from a fixed panel and from variants ClinVar already lists, so a gene where neither file shows a variant is less likely to be shared, not ruled out. Predictive carrier screening for couples is not currently offered in Finland[16].
References
- What are the different ways a genetic condition can be inherited?. MedlinePlus Genetics, US National Library of Medicine. Read 10 October 2026.
- Bell CJ, Dinwiddie DL, Miller NA, et al. Carrier testing for severe childhood recessive diseases by next-generation sequencing. Science Translational Medicine, 2011. doi:10.1126/scitranslmed.3001756
- Miller AC, Comellas AP, Hornick DB, et al. Cystic fibrosis carriers are at increased risk for a wide range of cystic fibrosis-related conditions. Proceedings of the National Academy of Sciences, 2020. doi:10.1073/pnas.1914912117 Chronic pancreatitis odds ratio 6.76 (95% CI 4.87 to 9.39); absolute risk 0.429 per 100 carriers.
- If a genetic disorder runs in my family, what are the chances that my children will have the condition?. MedlinePlus Genetics, US National Library of Medicine. Read 10 October 2026.
- Kandolin M, Pöyhönen M, Jakkula E. Estimation of carrier frequencies utilizing the gnomAD database for ACMG recommended carrier screening and Finnish disease heritage conditions in non-Finnish European, Finnish, and Ashkenazi Jewish populations. American Journal of Medical Genetics Part A, 2024. doi:10.1002/ajmg.a.63588
- Nelson DA, Deuster PA, Carter R, Hill OT, Wolcott VL, Kurina LM. Sickle cell trait, rhabdomyolysis, and mortality among U.S. Army soldiers. New England Journal of Medicine, 2016. doi:10.1056/NEJMoa1516257 Exertional rhabdomyolysis hazard ratio 1.54 (95% CI 1.12 to 2.12); death hazard ratio 0.99 (0.46 to 2.13).
- Naik RP, Derebail VK, Grams ME, et al. Association of sickle cell trait with chronic kidney disease and albuminuria in African Americans. JAMA, 2014. doi:10.1001/jama.2014.15063 Chronic kidney disease odds ratio 1.57 (95% CI 1.34 to 1.84); absolute risk difference 7.6%.
- Sidransky E, Nalls MA, Aasly JO, et al. Multicenter analysis of glucocerebrosidase mutations in Parkinson's disease. New England Journal of Medicine, 2009. doi:10.1056/NEJMoa0901281
- Anheim M, Elbaz A, Lesage S, et al. Penetrance of Parkinson disease in glucocerebrosidase gene mutation carriers. Neurology, 2012. doi:10.1212/WNL.0b013e318245f476 Penetrance 7.6%, 13.7%, 21.4% and 29.7% at 50, 60, 70 and 80 years.
- Rana HQ, Balwani M, Bier L, Alcalay RN. Age-specific Parkinson disease risk in GBA mutation carriers: information for genetic counseling. Genetics in Medicine, 2013. doi:10.1038/gim.2012.107 Cumulative risk among obligate-carrier parents 2.2% by 65 and 10.9% by 85.
- Martiskainen H, Willman RM, Harju P, et al. Monoallelic TYROBP deletion is a novel risk factor for Alzheimer's disease. Molecular Neurodegeneration, 2025. doi:10.1186/s13024-025-00830-3 Table 3: Alzheimer's disease hazard ratio 2.59 (95% CI 2.16 to 3.11), adjusted for APOE e4 and sex.
- Chu CS, Bancone G, Moore KA, et al. Haemolysis in G6PD heterozygous females treated with primaquine for Plasmodium vivax malaria: a nested cohort in a trial of radical curative regimens. PLOS Medicine, 2017. doi:10.1371/journal.pmed.1002224
- Uusimaa J, Kettunen J, Varilo T, et al. The Finnish genetic heritage in 2022: from diagnosis to translational research. Disease Models & Mechanisms, 2022. doi:10.1242/dmm.049490
- Taylor SM, Parobek CM, Fairhurst RM. Haemoglobinopathies and the clinical epidemiology of malaria: a systematic review and meta-analysis. Lancet Infectious Diseases, 2012. doi:10.1016/S1473-3099(12)70055-5
- gnomAD v4: genome callset, allele counts by genetic ancestry group. Genome Aggregation Database, Broad Institute, 2023. Read through the gnomAD browser API (dataset gnomad_r4, genomes) on 10 October 2026. Allele count / allele number, Finnish, non-Finnish European, African, Ashkenazi Jewish, East Asian: CFTR 7-117559590-ATCT-A 16/10612, 972/67954, 108/41556, 14/3472, 0/5180; HBB 11-5227002-T-A 0/10610, 11/68030, 1810/41554, 0/3470, 0/5186; HEXA 15-72346579-G-GGATA 1/10616, 20/68022, 2/41432, 44/3470, 0/5190; GBA1 1-155235843-T-C 11/10574, 124/67974, 8/41452, 101/3468, 0/5172; AGA 4-177438764-C-G 75/10602, 5/68030, 0/41552, 0/3470, 0/5184; SLC17A5 6-73644583-G-A 47/10606, 20/68004, 1/41566, 0/3470, 0/5184.
- Extensive analysis provides new insights into carrier risks for rare diseases. HUS Helsinki University Hospital, press release, 2024. Published 1 November 2024.
- Weedon MN, Jackson L, Harrison JW, et al. Use of SNP chips to detect rare pathogenic variants: retrospective, population based diagnostic evaluation. BMJ, 2021. doi:10.1136/bmj.n214
- Carrier Status reports. 23andMe Europe. Read 10 October 2026.
Last reviewed . Every number on this page links to the source it comes from; if one of them has moved, tell us.