Skip to content

What an Epigenetics Testing Kit Costs, and What You Get for the Money

Oak
A feathered reptile specimen on black glass, its iridescent plumage lined with tiny beads, dense and dull on one flank, sparse on the other.

A consumer epigenetics testing kit costs between about $200 and $600 for a single time point, and most vendors push you toward a subscription of two to four tests per year at $150 to $250 per test. The underlying measurement is DNA methylation, usually on an Illumina array covering 850,000 or roughly 935,000 CpG sites, and the array chemistry itself is only $150 to $300 of that price. The rest is the collection kit, extraction, and bisulfite conversion. The report and the provider’s margin account for the remainder. Research-grade methylation on the same platform, ordered through a core facility or a commercial sequencing provider, typically runs $300 to $800 per sample including library prep and QC.

The more useful question is what you get. You get an estimate of biological age, a pace-of-aging score, and some derived cell-type fractions. The measurement is reproducible enough to be interesting and noisy enough that a single test tells you very little.

The price breakdown by method

There are four ways to measure methylation, and they differ by an order of magnitude in cost.

The Illumina Infinium MethylationEPIC array (v1 at 850K probes, v2 at about 935K) is the default. It is what nearly every published epigenetic clock was trained on, which is the main reason to use it. Chip cost plus bisulfite conversion and hybridization lands around $250 to $400 per sample at volume, higher for single samples. You get IDAT files out, two per sample (Grn and Red channels), about 8 MB each.

Targeted assays are cheaper and less flexible. Pyrosequencing or amplicon sequencing of a few dozen CpGs can estimate age for under $50 in reagents. A simplified assay using eight CpGs by pyrosequencing reached a mean absolute deviation from chronological age of around 5 years in whole blood.1 Targeted bisulfite amplicon sequencing on a small panel has been pushed to similar accuracy with better scalability.23 The tradeoff is that you can only compute the clocks the panel was designed for.

Whole-genome bisulfite sequencing gives you every CpG in the genome at whatever depth you pay for, but 30x WGBS costs well over $1,000 and the per-CpG noise at that depth is worse than an array for clock purposes. Enzymatic conversion methods (EM-seq) avoid the DNA degradation that bisulfite causes, which matters if your input is low.

Pooled approaches are where the cost curve is moving. TIME-seq uses transposase-based barcoding to multiplex many samples in a single targeted methylation library, cutting per-sample cost by one to two orders of magnitude while still supporting clock construction.4 That work points at a near future where a methylation time point costs tens of dollars, which changes the calculus on how often you sample.

What the vendors are selling

The $500 consumer kit is an EPIC array plus a proprietary clock. The clock is the product, and it is the part you cannot inspect.

Several published clocks can be computed yourself from array data. These include Horvath’s multi-tissue clock, Hannum, and PhenoAge. GrimAge and DunedinPACE are also in this group. DunedinPACE is a rate estimate rather than an age estimate, trained on longitudinal decline in organ-system biomarkers. A newer pace estimator trained in the Framingham cohort follows the same design.5 Vendor clocks are often variants of these with retraining on internal cohorts, and the retraining is rarely documented well enough to reproduce.

Our view: pay for the raw IDAT files and compute clocks yourself. If a vendor will not hand over IDATs, treat the report as a black box and price it accordingly. Ask before you buy, in writing.

Reliability, which is the real cost question

Technical replicates of the same blood draw on the same platform disagree. For first-generation clocks the disagreement between replicates is commonly on the order of 2 to 4 years, which is similar in magnitude to the differences people get excited about between time points. Principal-component versions of the clocks (PC-Horvath, PC-PhenoAge, PC-GrimAge) were built specifically to suppress this noise and should be your default when you have array data.

Interpretation is contested in ways the marketing does not convey. A recent assessment of how these clocks behave under scrutiny concluded that their construction and validation still have unresolved problems for individual-level use.6 Cross-sectional validation work finds that methylation-based aging biomarkers track chronological age and some disease correlates reasonably at the cohort level.7 Cohort-level correlation is not the same as a stable individual readout.

Responsiveness to intervention is the claim most kits imply. A systematic look at how epigenetic aging biomarkers move in human intervention trials found that responsiveness varies substantially across clocks and across interventions, with pace measures and second-generation clocks behaving differently from first-generation ones.8 If you are running an n-of-1 experiment, choose your clock before you start and commit to it.

A protocol we would use

If you want a methylation time series that is worth analyzing, the design matters more than the vendor.

Sample from the same tissue and at the same time of day, in the same fasted state, every time. Whole blood is standard and is what most clocks were trained on. Saliva and buccal swabs give different cell mixtures and different clock outputs, and mixing tissues across time points makes the series uninterpretable.

Space time points at least six months apart unless you are testing a specific intervention with a defined window. Quarterly subscriptions mostly sell you noise. Collect at least three points before you try to read a trend.

Process all samples together if you can. Batch effects between array chips are real and comparable in size to biological effects. Freezing buffy coat or extracted DNA at -80C and running a batch of three later is better than three separate runs a year apart. Not all vendors support this.

Working with the files

The pipeline is short and the tooling is mature. In R, with minfi and sesame:

library(minfi); library(sesame)
rgset <- read.metharray.exp(base = "idats/", extended = TRUE)
# QC: detection p-values, drop probes p > 0.01 in >1% of samples
detp  <- detectionP(rgset)
# Normalization: noob + BMIQ, or funnorm for large heterogeneous sets
gmset <- preprocessNoob(rgset)
beta  <- getBeta(gmset)

Then filter cross-reactive probes and probes overlapping common SNPs, and drop sex chromosomes unless you need them. Estimate cell composition with estimateCellCounts2() on a blood reference, because leukocyte fractions drive a large share of apparent methylation change. Report clock outputs alongside the cell fractions below, or you will misread a shift in white-cell mix as aging.

  • CD8T
  • CD4T
  • NK
  • Bcell
  • Mono
  • Neu

For clock computation, methylclock and the dnaMethyAge package cover most published models. The PC clocks ship as a separate script and weight set from the Levine lab. Check how many required probes are missing on EPIC v2, since probe content changed between versions and naive imputation to the mean shifts clock estimates by a year or more.

Where a clinician belongs

Epigenetic scores are not diagnostic. Some vendors now market methylation-based dementia risk estimates, and survey work shows strong public interest in such testing along with clear concerns about what people would do with a result.9 Aging measures have also been associated with psychiatric outcomes in cohort studies, which is a population finding and not a personal prediction.10

If a result worries you, or if you are weighing a medical decision, take the data to a physician or a genetic counselor. We do not interpret these scores as health advice, and neither should a kit.

Questions people also ask

How much does epigenetic testing cost?

Consumer kits are $200 to $600 per time point, with subscription pricing bringing the per-test number to $150 to $250. Research-grade EPIC array methylation through a sequencing provider is $300 to $800 per sample. Targeted panels covering a few dozen CpGs cost far less in reagents but are usually not sold direct to consumers.

Are epigenetic tests worth it?

They are worth it if you will collect several time points under matched conditions and analyze the raw data yourself. A single test at one point in time gives you a number with a several-year uncertainty band and no baseline to compare it to.

What is the best epigenetic testing kit?

The best kit is whichever one returns raw IDAT files and lets you control collection conditions. Feature-by-feature, the arrays are the same hardware across vendors, so the differentiator is data access and batch handling rather than the report.

How do you test epigenetics?

Blood or saliva DNA is either bisulfite-converted or enzymatically converted, which turns unmethylated cytosine into uracil while leaving methylated cytosine intact. That difference is then read out by array hybridization or sequencing, giving a methylation fraction (beta value) between 0 and 1 per CpG site.

How much does a home DNA test kit cost?

A genotyping array kit for ancestry and common variants is $50 to $200 and measures DNA sequence. It does not measure methylation. Whole-genome sequencing at 30x now runs a few hundred dollars. These answer different questions from a methylation test and do not substitute for one.

Footnotes

  1. Laura Vidal-Bralo, Yolanda Lopez-Golan, Antonio Gonzalez. Simplified Assay for Epigenetic Age Estimation in Whole Blood of Adults. Frontiers in Genetics, 2016. https://doi.org/10.3389/fgene.2016.00126 ↩

  2. Benjamin Mayne, David Chandler, Christopher Noune, et al. Increased scalability and sequencing quality of an epigenetic age prediction assay. PLOS ONE, 2024. https://doi.org/10.1371/journal.pone.0297006 ↩

  3. Yang Han, Julia Franzen, Thomas Stiehl, et al. New targeted approaches for epigenetic age predictions. BMC Biology, 2020. https://doi.org/10.1186/s12915-020-00807-2 ↩

  4. Patrick T. Griffin, Alice E. Kane, Alexandre Trapp, et al. TIME-seq reduces time and cost of DNA methylation measurement for epigenetic clock construction. Nature Aging, 2024. https://doi.org/10.1038/s43587-023-00555-2 ↩

  5. William T Marella, Calen P Ryan, David Corcoran, et al. An epigenetic speedometer to measure Pace of Aging: FraminghamPACE. 2026. https://doi.org/10.64898/2026.07.07.26357388 ↩

  6. Steve Horvath. Putting epigenetic aging clocks on trial. Nature Medicine, 2026. https://doi.org/10.1038/s41591-026-04524-1 ↩

  7. Songnian Hu, Chao Ying, Jinhua Xue, et al. Validation of DNA methylation biomarkers for aging and disease research: A cross-sectional observational study. Aging Advances, 2026. https://doi.org/10.4103/agingadv.agingadv-d-25-00039 ↩

  8. Raghav Sehgal, Daniel Borrus, Jenel F. Armstrong, et al. Responsiveness of epigenetic aging biomarkers to longevity interventions in humans. Nature Medicine, 2026. https://doi.org/10.1038/s41591-026-04562-9 ↩

  9. Deirdre M. O’Shea, Devi Dhanekula, Swati Kumar, et al. Community perspectives on epigenetic dementia risk testing: Willingness, implementation preferences, and reasons for not testing in midlife and older adults. Alzheimer’s & Dementia, 2026. https://doi.org/10.1002/alz.71094 ↩

  10. Zoe Hart, Anna Großbach, Andrew J Simpkin, et al. The association between epigenetic ageing from childhood to early adulthood and psychotic-like experiences in early adulthood. Psychological Medicine, 2025. https://doi.org/10.1017/s003329172510055x ↩

Pothos builds longitudinal molecular profiles of individuals: whole-genome sequencing, single-cell RNA sequencing, and plasma proteomics, delivered as a detailed report, every data file, and stored cells. Build your profile.