G25 Studio: Beginner Manual

This guide is for people who are new to G25 coordinates and G25 Studio. It explains the ideas in plain language, walks through a simple first analysis, and points you to deeper documentation when you are ready.

Who this guide is for

You will get the most from this manual if you:

• Want to understand what G25 coordinates are before you interpret percentages.
• Plan to use G25 Studio to compare your DNA to reference populations and run ancestry models.
• Prefer a short, practical path first, then optional detail later.

You do not need a genetics background. You do need either your own coordinates or a way to obtain them (see Before you start).

What G25 coordinates are

G25 (Global 25) coordinates are a compact numeric description of your autosomal DNA in 25 dimensions. Each dimension captures part of how your genome relates to a large reference framework built with principal component analysis (PCA).

Think of coordinates as a position in genetic space, not as a story by themselves. They are the input you feed into tools like G25 Studio. Ethnicity percentages and admixture bars are outputs of modeling your coordinates against a chosen set of reference populations. They are an interpretation layer, not the same thing as the raw coordinates.

Scaled and unscaled

• Scaled coordinates weight dimensions by how much genetic variation they explain. They are the usual choice for ancestry modeling in G25 workflows.
• Unscaled coordinates keep the raw PCA-style scale. They can be useful for some research-style comparisons.

Use coordinates that match what your calculator expects (scaled vs unscaled). Mixing types by mistake usually leads to confusing or misleading results.

What coordinates are not

• They are not a medical diagnosis.
• They are not a genealogical record or proof of a specific ancestor.
• They are not a certificate of ethnicity or nationality.

Before you start

What you need

1. A DnaGenics account and a modern web browser with JavaScript enabled.
2. G25 coordinates for the sample you want to analyze (see below).
3. A few minutes for your first run. Heavier visualizations and large population sets can take longer.

How to get G25 coordinates

Typical paths include:

• Store or packs: Products that include G25 coordinate generation for your uploaded raw DNA (for example Starter Pack, Explorer Pack, or imputation services). See the Store for current options.
• Your processed samples: If you already have coordinates from a completed service, they may appear with your sample or download area.
• Reference-style coordinates: For learning, you can use coordinates from the library inside G25 Studio (not your personal genome).

If you only have raw DNA and no coordinates yet, finish coordinate generation before expecting meaningful G25 Studio results.

Coordinate line format

G25 Studio expects one line with a name followed by 25 comma-separated numbers:

MySample_scaled,0.0234,-0.0156,0.0089,-0.0312,0.0178,0.0123,-0.0098,0.0211,-0.0145,0.0067,-0.0189,0.0256,-0.0112,0.0198,-0.0076,0.0143,-0.0221,0.0095,-0.0167,0.0134,-0.0109,0.0182,-0.0126,0.0078,-0.0094

The exact numbers will differ for every person. The important part is: one label, then 25 values, separated by commas.

Three ways to load coordinates in G25 Studio

On the main G25 Studio screen you can:

1. Paste a coordinate line into the input area.
2. Use From Library to pick curated reference coordinates.
3. Use From Samples to pull coordinates tied to your uploaded samples (when available).

These labels match the buttons in the app so you can follow along while you work.

How G25 Studio works

G25 Studio is an ancestry analysis environment: you supply coordinates, pick a calculator, adjust configuration, run the analysis, then read results.

At a high level:

Coordinates -> Calculator -> Configure -> Analyze -> Results

What is a calculator?

A calculator is a named bundle of reference populations (ancient, modern, or both) plus metadata (for example scaled vs unscaled, era, author). Your coordinates are compared to those references to produce distances, admixture-style percentages, and optional charts.

Different calculators give different results because each one only "sees" the populations it contains. A calculator focused on one world region often gives more detail there. A broad global calculator gives a wider but sometimes less granular picture. Neither is automatically "right"; they answer related questions with different tools.

Why calculator choice matters

Your genome is modeled as a mixture of the populations the calculator knows. If the calculator does not include good references for part of your ancestry, that part may be spread across other populations or show a weaker fit. That is a limitation of the reference panel, not a judgment about your DNA.

Your first beginner analysis

Follow this path the first time. You can add complexity later.

1. Load coordinates (paste, From Library, or From Samples).
2. Open Browse Calculators and pick a calculator that matches your question (for example a regional calculator if you know roughly where your recent ancestry is from, or a broader one to start).
3. Open the configuration area and set:
   • Algorithm: Montecarlo V1 (standard, friendly default).
   • Distance: Euclidean (good general-purpose choice).
   • Maximum populations: about 5 for easier reading.
   • Group populations: On if you want a more regional summary.
   • Ancestors Predictions (Oracle): Off for the very first run (add it when you want multi-way "what if" scenarios).
   • Analysis tools: enable PCA only for a quick visual anchor.
4. Click Calculate Results and wait until the run finishes.

This pattern mirrors the in-app Configuration Guide recommendations for a first analysis.

How to read the results

Admixture composition (Monte Carlo style)

The tool tries to explain your coordinates as a blend of reference populations with flexible percentages. Higher percentages mean stronger statistical similarity in that calculator's space. They do not mean you have a recent ancestor from every labeled group.

Fit score

The fit score measures how well that calculator's reference set can reconstruct your coordinates. Lower is better.

Many reports treat very low scores as excellent fits and moderately low scores as still useful, but what is achievable depends on the calculator. A higher fit score often means "this panel does not capture all of your ancestry well," not that your DNA is defective. Try another calculator with references that better match your background.

Closest populations (often linked to Oracle-style output)

Closest populations highlight references whose coordinates sit nearest to yours in G25 space. That is a neighbor idea. It is not the same as the mixture percentages, which optimize a blend. Both are useful; they answer slightly different questions.

Ancestry timeline (when shown)

If present, a timeline is a statistical estimate of when components may have entered your pedigree. Treat it as a guide, not a dated family tree.

Free vs PRO

Some algorithms, distance metrics, population customization, extra visualizations, and extended Oracle result counts are PRO features. If a control is locked or labeled as requiring an upgrade, you can still learn the workflow with Montecarlo V1, Euclidean distance, and PCA.

Upgrade when you want to experiment with more optimizers, more charts, or stricter population controls. 

Monte Carlo vs Oracle

In G25 Studio, Oracle-style output is configured under Ancestors Predictions (Oracle).

• Monte Carlo (admixture-style modeling): searches for a flexible combination of populations and percentages that fits your coordinates well. Good for "what mixture explains me in this calculator?"
• Oracle (Ancestors Predictions): explores fixed share scenarios (for example equal splits across two, three, or four populations) to help you think about parent and grandparent style combinations. Good for "what if my ancestry were split evenly in these ways?"

Use them together after you are comfortable: Monte Carlo for the main mixture read, Oracle for structured "what if" checks.

Glossary

Definitions are short on purpose. Terms appear in alphabetical order.

Admixture composition
The modeled breakdown of your genome into shares of reference populations for a chosen calculator.

Ancestry modeling
Using your G25 coordinates and a reference panel to estimate population-like contributions and related summaries (not the same as building a documentary family tree).

Ancestors Predictions (Oracle)
Optional analysis that scores fixed-share combinations of populations.

Calculator
A named reference panel and settings bundle used to interpret your coordinates.

Closest populations
Reference populations with the smallest genetic distance to your coordinates in G25 space for the chosen method.

Distance metric
The mathematical rule used to measure how far your coordinates are from a reference population (for example Euclidean).

Euclidean distance
Straight-line distance in the 25-dimensional coordinate space. A common default.

Fit score
A summary of how well the calculator's references can match your coordinates. Lower is better, but good ranges depend on the panel.

G25 coordinates
Twenty-five numeric values (plus a sample label) describing your position in the Global25 PCA-style framework.

Genetic distance
How far apart two coordinate sets are under the chosen distance metric. Smaller usually means more similar for that metric.

Global25 (G25)
The 25-dimensional coordinate system used across G25 Studio and compatible tools.

Monte Carlo (Montecarlo)
An iterative optimization approach used to search for strong mixture fits under your chosen settings.

PCA (Principal Component Analysis)
A visualization that projects coordinates onto the main axes of variation so you can see how close you sit to reference populations.

Reference populations
Named genetic averages (ancient or modern) that define what a calculator can model.

Scaled coordinates
Eigenvalue-weighted coordinates; the usual default for G25-style modeling.

Unscaled coordinates
Coordinates on the raw PCA scale; use when your workflow intentionally calls for them.

Common beginner questions

Why do I see regions I have no family stories from?
Population genetics does not follow modern borders. Small shares can reflect deep shared ancestry with a reference group, not a recent relative from that label.

Why do two calculators disagree?
Each calculator uses a different reference set. Different references partition your genome differently. Compare calculators to learn robust patterns, not to pick a single "true" pie chart.

Can G25 prove my ethnicity?
No. It is a statistical comparison framework. Documentary genealogy and appropriate genetic genealogy tests serve different purposes.

Troubleshooting

Next steps

• Explore populations tied to your calculators: G25 Population Library.
• Browse calculators and metadata on the public G25 Calculators areas of the site when you want to compare panels before you run.

Community help is linked from the main user guide if you want to discuss results with other users.

Disclaimer

G25 Studio is intended for education and research-style exploration. It is not a medical product. Interpret results with appropriate caution and avoid over-claiming about identity or family history based on modeled percentages alone.