DomDNA Learn · Published Lab articles
Health questions.
Clearer answers.
Make sense of the health claims you keep seeing. We explain what the evidence says, where it is uncertain, and what you can ask next.
Berberine and 'natural Ozempic': how to read the claims
Berberine is marketed for weight loss, but the evidence and safety questions differ from prescription medicines. Check the claim before buying.
Perimenopause brain fog and sleep: what to bring to your GP
Brain fog, broken sleep or changing periods? Use a short symptom record to prepare for a GP conversation without relying on a wellness score.
Vitamin D in autumn: a clear guide for UK adults
As autumn arrives, check the NHS vitamin D guidance, the limits of symptom guessing and the information to take to a pharmacist or GP.
AlphaGenome Atlas: what it means for your DNA report
AlphaGenome Atlas predicts molecular effects of DNA changes. Understand the research advance, its limits and why it cannot diagnose you from a consumer file.
Biological-age tests: what a younger score really means
Can a biological-age test prove you are ageing more slowly? Learn what methylation clocks estimate and why a September 2026 preprint adds caution.
Your sleep score is an estimate: how to use it well
A sleep tracker can help you notice routines, but its score needs context. Learn what to check, what to record and when to seek professional advice.
Magnesium glycinate for sleep: check the claim before the label
Does magnesium glycinate help sleep? Understand the limited evidence, supplement labels and safety questions before adding another bedtime product.
CGM without diabetes: what a glucose graph can tell you
Thinking about a continuous glucose monitor without diabetes? Understand the sensor, its limits and the questions to ask before paying for more data.
Creatine for brain fog: what the research can tell you
Creatine brain-health claims need context. See what a small sleep-deprivation trial found, what remains uncertain and how to read supplement claims.
GLP-1 weight loss: keeping muscle and nutrition in the picture
Eating less on a GLP-1 medicine? Learn why strength, food intake and support matter, and what to discuss with your prescriber or dietitian.
Fibremaxxing: how to eat more fibre without overdoing it
Curious about fibremaxxing? Start with the UK fibre target, everyday food swaps and a gradual approach that gives your gut time to adjust.
Research library · 112 resources
Released 4 October 2026. AI-assisted educational articles, not independently clinically reviewed. Original research dates, source-access limits and historical preview wording remain visible.
Your blood-pressure cuff is part of the measurement
Why cuff fit matters, what a randomised trial found, and how to keep a home blood-pressure record your healthcare team can interpret.
Potassium salt substitutes: the safety question behind the swap
What the salt-substitution trial tested, who WHO guidance excludes, and why the product label and household context matter.
Thinking of cutting gluten? Ask about coeliac testing first
Why dietary changes can affect coeliac testing, what genetic susceptibility cannot confirm, and how to prepare a useful appointment note.
Taking metformin? Put vitamin B12 on your review list
The UK monitoring advice, the limits of a long-term trial, and what to bring to a B12 discussion without changing treatment yourself.
When an HbA1c result needs a second look
How red-cell lifespan and haemoglobin variants can affect HbA1c interpretation, and what to ask when results appear inconsistent.
Snoring and daytime sleepiness deserve a proper conversation
What to tell a GP about possible sleep apnoea, what a sleep study adds, and why treatment claims need a named outcome.
Ferritin needs context before an iron shopping list
Why ferritin may need other clinical information, why the cause of deficiency matters, and how to avoid universal supplement targets.
Before comparing HRV numbers, check what was measured
Why HRV labels and measurement windows matter, what a six-device study can establish, and how to preserve useful source labels.
Taking biotin? Put the bottle on your blood-test checklist
Biotin can affect some laboratory assays. Learn what to record before a blood test and why preparation instructions need to come from your clinical team.
Your eGFR result needs the test method beside the number
An eGFR value is an estimate. Keep the method, sample date and accompanying results together, then ask how they fit your clinical assessment and follow-up.
Lipoprotein(a): bring the actual result to the risk conversation
Lp(a) is distinct from a usual cholesterol panel. Bring the actual report, units and family-history questions to a discussion about overall cardiovascular risk.
A watch rhythm alert: save the record, check the symptoms
A watch alert needs its exact feature name and symptom context. Learn what to save, what the device cannot establish and when to seek urgent help.
A genetic variant of uncertain significance needs a follow-up plan
A variant of uncertain significance leaves a clinical question unresolved. Keep the original wording and ask who explains the result and any future updates.
A negative carrier screen still has a defined scope
A negative carrier screen has limits. Find the conditions and variants assessed, preserve family-history uncertainty and ask about the risk that remains.
Build a family health history without filling the gaps with guesses
Make family-history notes that distinguish confirmed information from recollection, respect relatives' choices and support a useful healthcare conversation.
An oxygen reading needs symptoms and device limits beside it
Pulse oximeters estimate oxygen saturation. Keep symptoms, device instructions and any clinical plan alongside the reading, without inventing a correction.
Your sleep-stage chart and a clinical sleep test answer different questions
Understand what sleep-stage labels describe, how clinical tests collect different signals and what to take to an appointment about a sleep concern.
Seven hours can sit on very different sleep schedules
Record sleep timing alongside duration without treating a regularity score as a lifespan forecast. Use context to choose one workable routine change.
Logging sleep after a night shift: keep both dates
Keep shift and sleep timestamps together across midnight. Record naps and the comparison window so an app's date label does not confuse your notes.
A strange workout pulse reading needs the sensor conditions beside it
Save fit, activity and symptoms beside an odd workout pulse reading. Understand optical sensor limits without guessing a cause or changing treatment.
A missing ring night belongs in the record as missing
Record the missing fields and usable nights before comparing ring data. Check manufacturer instructions without replacing gaps with invented values.
A wearable temperature deviation needs its baseline and context
Read temperature deviations with their metric and baseline. Keep thermometer checks and fertility decisions separate from an unexplained wearable change.
Choose a step goal from your routine, not a magic 10,000
Choose a workable step goal from your recorded routine. Read mortality associations with their limits and keep other forms of activity in the picture.
Your wearable VO2 max estimate and a lab measurement need separate labels
Label wearable VO2 max estimates with their method, date and units. Understand lab testing and validation limits before interpreting a changing value.
A genetic risk headline needs a starting number
A 50% increase needs a baseline, timeframe and outcome. Learn to separate absolute risk, relative risk and odds before interpreting a genetic headline.
A polygenic score needs evidence from the people using it
A polygenic score needs population-specific validation, not just a large development sample. Ask which score version was evaluated and in whom.
A GWAS association leaves more than one question open
A GWAS marker can be associated with a measured trait without proving a mechanism or predicting your outcome. Read the endpoint before the gene headline.
Different DNA coordinates can refer to the same region
Two DNA reports can use different reference assemblies and coordinates. Keep the reference version and notation before deciding their results disagree.
Two SNP results can leave their chromosome pairing unresolved
Two SNP genotypes do not always reveal which variants share a chromosome copy. Learn what phasing adds and what an unphased report leaves unresolved.
Penetrance and expressivity answer different questions
Penetrance asks how often a feature appears; expressivity asks how it varies when present. Keep age, outcome and study population with the estimate.
A genetic report’s date tells you which explanation you received
A later report date does not explain what changed. Separate sample measurement, analysis scope and evidence versions before assuming new genetic findings.
Compare a genetic panel’s scope before counting its genes
More genes do not guarantee broader detection. Compare targets, variant types and assay limits, and keep an unresolved no-call separate from a negative result.
Give your evening a workable wind-down
Explore evening wind down routine with a practical everyday example, relevant research limits and one workable next step. General education, not a personal.
Add breakfast protein without turning one meal into a rule
Explore breakfast protein daily intake with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Creatine and a heavier scale: keep the training question in view
Explore creatine weight gain water with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Put a movement break inside the working day you have
Explore movement breaks desk work with a practical everyday example, relevant research limits and one workable next step. General education, not a personal.
Try one manageable walking leg of your usual journey
Explore walking commute small change with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Keep frozen vegetables for the evening you have no plan
Explore frozen vegetables easy meals with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Plan the fish meal before treating capsules as its replacement
Explore fish meals versus omega 3 supplements with a practical everyday example, relevant research limits and one workable next step. General education,.
Make two strength sessions comparable before calling it progress
Explore comparing strength training sessions with a practical everyday example, relevant research limits and one workable next step. General education, not.
Leave room between demanding strength sessions
Explore rest days between strength workouts with a practical everyday example, relevant research limits and one workable next step. General education, not.
Give daylight a place in your daily routine
Explore daylight sleep routine outdoors with a practical everyday example, relevant research limits and one workable next step. General education, not a.
A sleepy evening is not the same as better sleep
Explore alcohol sleepy sleep quality with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Plan endurance drinks around the session, not a shopping list
Explore endurance hydration planning with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Find a calcium food that fits the meals you already eat
Explore calcium foods fortified alternatives with a practical everyday example, relevant research limits and one workable next step. General education, not.
Make home strength training easy to return to
Explore home strength training consistency with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Give walking and strength training different jobs in your week
Explore balance aerobic strength activity with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Do not use soreness as your workout score
Explore soreness workout quality progress with a practical everyday example, relevant research limits and one workable next step. General education, not a.
Why TSH and free T4 belong in the same conversation
Learn why TSH and free T4 describe different parts of thyroid feedback, why panels differ and which question to ask about a report. General education.
Does a triglyceride result need a fasting sample?
Understand why meal context matters without dismissing every non-fasting lipid sample. Check the appointment instructions and keep the full report.
Liver enzymes are not a single liver-function score
Read ALT and ALP as distinct measurements, not a liver-performance percentage. Explore why the full panel and laboratory comments matter for discussion.
A reference interval is not always a decision threshold
Explore the difference between a population reference interval and a clinical decision limit, using fictional numbers rather than personal thresholds.
How much does a change between two lab results mean?
Understand why a line between two laboratory results is not yet a clinical trend. Keep comparable reports and ask how expected variation is assessed.
Preparing for a blood test without inventing a fasting rule
Find the preparation instructions for your actual blood collection, clarify medicine or meal conflicts and report deviations honestly. General education.
A urine dipstick and a culture answer different questions
Learn why a quick urinalysis and a later culture provide different information, why symptoms matter and how to ask which results remain pending.
What the small red-cell labels on a blood count describe
Understand how red-cell number, haemoglobin and average cell size differ, and why these descriptions do not identify a deficiency on their own.
Why CRP and ESR need not tell the same story
Learn why CRP and ESR are not interchangeable inflammation scores, how test names and timing matter and what to ask about an unexpected pattern.
When a potassium result also carries a sample warning
Understand why potassium sample comments matter without assuming a result is harmless. Ask the requester how the warning affects the planned next step.
Total, adjusted and ionised calcium are different labels
Read the full blood-calcium label, distinguish measured from derived results and avoid treating calcium as a dietary-intake or bone-strength score.
Why blood albumin is not a score for protein intake
Understand why albumin and prealbumin cannot stand in for dietary or muscle assessment, and bring actual eating difficulties into the clinical conversation.
Why urine albumin is often reported as a ratio
Explore what the urine ACR ratio describes, why confirmation can be requested and why urine and blood kidney measurements answer different questions.
Why a screening FIT does not settle a new bowel symptom
Understand why routine bowel screening and symptom investigation are different pathways, and why a previous screening letter should not settle new symptoms.
Troponin results belong to a clinical timeline
Understand why troponin belongs with symptom timing, ECG findings and clinical assessment. Suspected heart attack needs emergency help, not portal scoring.
A sensitive test does not make every positive result certain
Use an invented test example to distinguish sensitivity, specificity and positive predictive value without turning performance claims into a diagnosis.
An imputed genotype is a prediction, not another measurement
Learn why an imputed DNA file contains predicted genotypes, what reference data contributes and which file details to preserve before comparing results.
An allele frequency counts copies, not people
Understand why allele frequency counts DNA copies rather than people, how available observations affect its denominator and why rarity is not a diagnosis.
More sequencing reads do not answer every quality question
See what sequencing read depth describes, why mapping and base quality still matter and which technical fields to retain without inventing certainty.
Why one DNA change can have several predicted consequences
Learn why a DNA change can receive different predicted consequence labels across transcripts and why keeping the transcript identifier helps comparisons.
A tumour result does not automatically describe inherited DNA
Understand why tumour testing does not automatically describe inherited DNA and how keeping the test purpose visible prevents misleading report comparisons.
Mosaicism changes the question you ask about a sample
Explore how mosaicism changes the meaning of a sample result, why tissues can differ and how to record specimen context without estimating personal risk.
A mitochondrial percentage needs its tissue beside it
Learn what mitochondrial heteroplasmy describes, why the tissue and testing method belong beside the percentage and why it is not a personal forecast.
A copy-number change is about how much DNA is present
Understand what a copy-number finding describes, how it differs from a sequence change and why a test's stated scope matters when comparing reports.
Heritability is not the genetic percentage of a person
Understand why heritability is not a genetic percentage of one person, why environment and studied populations matter and how to read research headlines.
Several nearby genetic signals may not be independent clues
Learn why nearby variants can carry correlated research signals, why counting them can overstate independence and what study methods need to stay visible.
A gene-expression result is not a list of DNA changes
Understand how gene-expression results differ from DNA sequence findings, why tissue and measurement context matter and what to retain when reading research.
ClinVar stars describe review status, not your disease probability
Understand what ClinVar review stars describe, why they are not personal disease probabilities and which submission details matter before quoting a record.
Secondary findings are a choice about a different testing question
Learn what secondary genomic findings mean, why testing choices and consent matter and why older guidance cannot serve as a current personal gene checklist.
A PubMed link does not tell you the whole publication status
Learn how to distinguish preprints from journal articles, check publication records and retain version context without treating indexing as an endorsement.
An ancestry estimate is a comparison, not a definition of identity
Understand why genetic ancestry estimates depend on reference comparisons, why they do not define identity and what context to retain when results change.
Reading the quiz preview without giving an email address
See how the current quiz preview allows results without an email, what that does and does not prove and how sharing differs from sending personal answers.
What the quiz preview means by a browser copy
Understand the preview's browser copy, its expiry and restoration rules and why local storage is neither a cross-device account nor guaranteed persistence.
A saved quiz and an accepted email are different receipts
Learn why the preview separates saving from email status, why accepted email is not proven delivery and what restored status can and cannot confirm.
Use a quiz focus as a question to observe, not a diagnosis
Use the preview's lifestyle focus as a practical observation prompt, while keeping it separate from diagnosis, genetic inference or supplement recommendations.
Saving a quiz and choosing updates should be separate decisions
Understand how the preview separates saving from optional marketing choices, why consent language matters and why code inspection is not legal certification.
Randomised is useful. How was the assignment hidden?
Randomised studies still need clear assignment methods. Learn why concealment before enrolment differs from blinding and how to note a reporting gap.
Who was blinded, and which outcome could expectations change?
Learn how participant, assessor and analyst knowledge can affect different outcomes, and why a double-blind label cannot answer every trial-quality question.
Who stayed in the analysis after people stopped participating?
Follow the participants from assignment to analysis. Learn why completer results, missing measurements and an intention-to-treat label need separate attention.
A crossover trial needs more than two different periods
Receiving both options does not erase time. Learn how order, lasting effects and period changes shape crossover research and a useful comparison note.
Why improvement after an unusually bad week is hard to interpret
An extreme starting week can make later improvement hard to attribute. Learn how selection, ordinary variation and context affect a before-and-after story.
Was the headline outcome chosen before the results were known?
Check whether the headline outcome was planned before results appeared. Learn why additional comparisons and outcome changes need visible interpretation.
A changed biomarker does not name the benefit on its own
A biological change can be useful without proving patient benefit. Separate the measurement, its response and the clinical claim it is used to support.
A confidence interval for the mean is not your personal range
A confidence interval around a group mean describes uncertainty in that estimate, not a personal outcome range. Learn which quantity the brackets belong to.
Would the measured difference matter outside the statistical test?
A significant score change is not automatically a useful benefit. Read the scale, effect size, uncertainty and context before judging practical importance.
Who volunteered before the study began?
Large studies can remain selected samples. Learn how recruitment affects prevalence and associations, and how to keep the participant population visible.
Today, last week or usually: the question changes the answer
Today, last week and usually can measure different experiences. Learn why question wording, recall periods and changed instructions belong beside a score.
Two devices can move together without giving the same readings
Two methods can track together while giving different readings. Learn why paired differences, units and intended use matter more than correlation alone.
Did your trend change when the software changed?
A wearable chart can span different firmware and algorithm versions. Learn how to record updates without assuming they caused a trend or inventing a correction.
What changed between the studies in the pooled result?
A pooled estimate does not erase differences between studies. Compare outcomes, populations and follow-up before deciding what a meta-analysis can answer.
No clear difference is not the same as demonstrated equivalence
No clear difference does not establish equality. Learn how non-inferiority and equivalence designs use stated margins, uncertainty and a specific outcome.
Read what the trial planned, not only what it reported
A trial identifier is not an endorsement. Compare the original outcome and dated record history with the publication to understand what was planned and changed.
A star allele is more than a highlighted SNP
Three labels can describe different levels of a medicine–gene result. Learn what to preserve before comparing reports or asking about tested coverage.
CYP2D6: three copies still leave a question
A total copy count does not identify which allele was duplicated. Learn which structural limitations to preserve in a CYP2D6 report.
Why a medicine list belongs beside a gene result
An inherited result does not update with a prescription change. Understand why current medicine context and report dates belong together.
A CYP2C19 result needs the clopidogrel indication
The reason a medicine is used matters to a pharmacogenomic interpretation. Keep the indication with a CYP2C19–clopidogrel report.
A negative DPYD screen is not a complete toxicity forecast
Keep the tested variants, test type and treatment context attached to a DPYD result. Negative screening is not a promise of no toxicity.
TPMT and NUDT15 are two questions, not two names for one test
Separate gene rows, test methods and coverage. An activity result and a genotype assignment are not interchangeable documents.
SLCO1B1 is not a verdict on every statin
Name the statin and the endpoint before interpreting a panel heading. Keep gene findings, symptoms and assessments separately recorded.
“HLA positive” leaves out the part you need
Preserve the full allele name and named medicine context. A generic HLA label cannot replace specific testing or urgent symptom advice.
UGT1A1 and atazanavir: name the bilirubin question
A bilirubin-related gene–medicine section is not a general liver-health score. Separate the report, measurements and symptom assessment.
Warfarin genes do not replace the INR record
Inherited information and current anticoagulation monitoring are different records. Keep both without calculating a dose from gene names.
CYP3A5 is not a tacrolimus blood level
Separate a genetic assignment from a measured concentration and the care team’s sampling instructions. No level or dose is inferred.
Before trusting a drug-panel colour, find its reference
Trace a drug-panel category to its named source, edition and endpoint. A colour is not a universal clinical verdict.
Earlier articles
From the archive. Check each article’s date and references; older material may not reflect current product capabilities or guidance.
23andMe bankruptcy
An archive guide to the 23andMe bankruptcy, what it raised about genetic-data ownership, and questions to ask about your information.
23andMe alternatives
An archive comparison of DNA testing services, with questions about privacy, data access and how to choose between providers.
Deleting genetic data
An archive guide to deleting genetic data, including account settings, sample storage and the limits of deletion requests.
Genetic privacy
Explore why genetic information is personal, what sharing it can mean for relatives, and the privacy questions worth asking.
DNA and supplements
An introduction to genetic differences and supplements, covering the questions behind personalised nutrition.
How genetics relates to medicines
Explore how genetic differences relate to the way medicines are processed, with topics to discuss with a prescriber.
The microbiome and DNA
An introduction to the relationship between gut microbes and genetics, and the questions researchers are exploring.
AI and genetic analysis
Explore how AI is used to analyse genetic information and the ideas behind turning large datasets into health insights.
Nutrigenomics
An introduction to nutrigenomics: the study of how genetic differences relate to food, nutrients and metabolism.
Genetics and healthy ageing
Explore the research themes linking genetics, ageing and everyday habits in this article from the archive.
Comparing DNA kits
An archive comparison of five DNA testing services, focusing on privacy, the information provided and how results are used.
Genetic-data sales
An archive look at commercial genetic-data partnerships and the questions they raise about consent and ownership.
The cost of genetic testing
Explore the business models behind low-cost genetic testing and the questions to ask about how your data may be used.