Read your report
Confidence Tiers explained: why your report grades evidence (Tier 1, 2 and 3)
Some lines in a genome report rest on decades of replicated science. Others are early, contested, or true only for some ancestries. A good report tells you which is which before you read a word of the finding.
The claim. If a result appears in my DNA report, it must be a proven fact about me. Every line on the page carries the same weight, so a finding near the bottom is as solid as a finding near the top.
Verdict. Not every genetic finding is equally certain, and a report that pretends otherwise is misleading you. Aimosti sorts results into three confidence tiers so a reader can see at a glance how much weight each finding can bear. The tier is part of the result, not decoration.
Genetic evidence comes in very different strengths. A pharmacogenomic gene-drug pairing backed by a clinical consortium is not in the same league as a single small study of a behaviour trait, even though both can be printed in the same font. The wider field already grades this difference: ClinVar rates each variant interpretation with review stars, CPIC assigns evidence levels to gene-drug pairs, and ClinGen scores how firmly a gene is linked to a disease. Aimosti restates those existing grades as three plain-language tiers rather than inventing a new scale.
What a confidence Tier is
A confidence tier is a label that says how much trust a finding has earned, before you even read what the finding is. It is about the strength of the evidence behind a result, not about how serious that result would be if it were certain. Aimosti uses three.
Tier 1 · Established covers strong, replicated science, often recognised in clinical guidelines: the most trustworthy signals in a report. Tier 2 · Well-supported covers solid, reproducible findings that are pharmacogenomic and carrier-grade, reliable, though not all are clinically actionable. Tier 3 · Emerging / Speculative covers readings that are real but preliminary: mixed, contested, or population-limited, interesting to explore and never to act on.
Why grade evidence at all, instead of stating facts
Printing every finding in the same plain sentence would flatten a real difference. It would set a gene-drug interaction that changes prescribing in many hospitals next to a single-study hunch about a personality trait, and let the reader assume both are settled. That is how consumer genetics earns its reputation for overpromising.
Grading is the better alternative. By marking where a finding sits, a report keeps evidence and effect separate: it can show an interesting but weak signal without implying it is proven, and it can give a strong, unglamorous finding the weight it deserves. The label carries the caution, so the prose does not have to repeat it on every line.
How the wider field already grades evidence
Aimosti did not invent tiered confidence. ClinVar, a public archive of variant interpretations, rates each one with zero to four review stars for how well-reviewed it is, so more stars mean more expert agreement. CPIC, the consortium that writes pharmacogenomic prescribing guidance, assigns evidence levels to gene-drug pairs. ClinGen scores how firmly a gene is tied to a disease, from definitive down to disputed.
Polygenic and exploratory readings sit lower for a concrete reason. Genome-wide studies use a strict statistical bar, a p-value at or below 5×10⁻⁸, before a signal counts as unlikely to be a fluke. Many real but small-effect variants sit just below that bar, which is exactly why they belong in a lower tier rather than among the established findings. Aimosti's three tiers are a plain-language roof over these existing systems, not a replacement for them.
What the tiers look like on a report
The public SAMPLE report at /sample shows all three tiers at once. It is shown for illustration and is not a real genome, so no number on it belongs to anyone. Clinical findings appear under the Tier 1 · Established heading. Pharmacogenomic and carrier results, along with the polygenic scores, carry a Tier 2 · Well-supported chip; a polygenic score is one that adds up many small-effect variants into a single statistical tendency, which is why it is well-supported rather than established.
Lower down, a divider reading "credibility firewall · exploratory below" marks the start of Tier 3, where the Frontier and Fringe readings live. The same finding never moves between tiers to look more or less certain: its tier is fixed by the evidence, and the chip on each card repeats the grade so it travels with the result wherever the result appears.
What a tier does not tell you
A tier grades the evidence, not you. A Tier 1 finding is still a risk factor or a carrier status, not a diagnosis, and a high tier never means the result is severe or certain for any one person. The strength of the science and the meaning for your health are two different questions, and the tier answers only the first.
A Tier 3 reading works the other way: it can be genuinely interesting and still wrong for you, because the evidence is thin or was gathered in a different population. The right reading of a low tier is curiosity, not action. A tendency is not a verdict, and the label is the point.
What Aimosti would (and wouldn't) show you
On a report, every finding sits inside one of three tiers, each introduced by a one-line banner that says what the tier means, and each card carries a small confidence chip repeating its tier. A divider labelled a credibility firewall separates the exploratory tier from the rest, so the speculative readings can never be mistaken for the established ones. Nothing is hidden: the weakest findings are still shown, just clearly marked.
What we won't claim
We won't dress a Tier 3 reading up as a Tier 1 fact, drop the tier label to make the report look more certain, or let a confident visual design imply confident science. We restate the evidence grades that bodies like ClinVar and CPIC publish; we don't upgrade them. The tier is part of the answer, and removing it to look more impressive would be misleading.
Bottom line. Not every genetic finding is equally certain, and a report that pretends otherwise is misleading you. Aimosti sorts results into three confidence tiers so a reader can see at a glance how much weight each finding can bear. The tier is part of the result, not decoration.
Related: Every module in your report, grouped by confidence tier. Restated from: ClinVar (NCBI): variant review-status stars · CPIC: Clinical Pharmacogenetics Implementation Consortium · ClinGen: gene-disease validity classification · GWAS Catalog (EMBL-EBI / NHGRI).