Advanced Segment Mapping
This hub is about the working method: how to record DNA segments, connect them to your matches, and keep track of how well each ancestral hypothesis is supported. The theory behind segments, IBD and triangulation is covered in the Guide to DNA Triangulation, so this page stays practical.
A segment that overlaps between you and two or more matches is a good reason to look closer. It does not, on its own, establish a relationship or identify the ancestor. The tool and the workflow below help you organise evidence. They cannot make the following conclusions for you:
- That the overlapping people share the same ancestor. Several ancestors, endogamy or chance can produce the same pattern.
- That a short segment is genuinely shared by descent. Small segments are often false positives.
- That a match's tree is correct. A triangulated group only helps if the trees behind it are right.
- The exact relationship. Shared cM fits several relationships, so it supports ranges, not a single answer.
The workflow in five steps
Record each segment consistently
Log every segment you want to test with the same fields, so segments from different platforms can be compared later.
Field What to record Why it matters Chromosome 1–22 (and X, if relevant) Overlap only counts on the same chromosome. Start and end position Base-pair positions, e.g. 45,200,000 to 58,900,000 Needed to judge overlap precisely. Genome build GRCh36, GRCh37 or GRCh38 Positions from different builds are not directly comparable. Convert first. Length cM and, if shown, number of SNPs Length is the main guide to whether a segment is likely to be real. Platform and kit Where the segment was reported Each platform has its own matching and phasing method and thresholds. Date checked When you pulled the data Platforms update matching, so results can change. Link the segment to DNA matches
For each segment, list the matches who overlap it and note:
- Which platform each match is on, and whether the matches also match each other on that segment.
- The total shared cM and the estimated relationship range.
- Whether the match has a tree, and which ancestors it names.
- Whether the match is known to come from your maternal or paternal side (for example through a parent or a known relative).
Give every hypothesis a status
Write the hypothesis as a sentence ("This segment comes from couple X") and give it one of the levels below. Raise or lower the level when new evidence arrives. There is deliberately no "proven" level.
1. LeadOverlap seen. Trees not yet compared, or only a small segment.
2. TentativePlausible shared ancestor in the trees, but based on one match or a short segment, or with unresolved doubts.
3. SupportedSeveral matches who also match each other converge on the same ancestral couple, and alternatives have been considered.
4. Well supportedAs level 3, plus documentary records and, where possible, testing of additional relatives. No unresolved conflicts.
Levels 3 and 4 describe how much evidence you have. They are not a statement of certainty.
Do not mistake overlap for a settled relationship
Before you raise a status, check the usual traps:
- Pile-up regions: stretches where many people match, which can inflate apparent triangulation.
- Endogamy: the same ancestors appear many times in a population, so shared segments cannot be pinned to one line without extra evidence.
- Multiple shared ancestors: two matches can be related to you through different lines yet still overlap.
- Phasing and matching errors: especially for short segments and without parental or sibling data.
Document sources and alternative explanations
For each hypothesis, store the evidence and the competing explanations together. Record:
- Sources for the tree links: records, citations or the match's own tree, and how reliable each is.
- At least one alternative explanation and why it is less likely, or why it is still open.
- What would change your mind, and the next test or record that could settle it.
One segment, two possible readings
You have a segment on chromosome 7 (GRCh37, about 45.2 to 58.9 Mb, 14 cM). Match A and Match B overlap it, and A and B also match each other there. Their trees both lead to the same couple, born around 1840.
What you can say
Hypothesis: the segment descends from that couple. Status: Supported (level 3), because two matches converge and match each other.
What you cannot say yet
That the link is established. Both trees might share an error, the pair might be related through a second line, or the area might be a pile-up region.
Next step: look for a third match from a different branch of the couple's descendants, check the region against known pile-up areas, and find a record that confirms the tree links. If it holds, move the status to level 4.
Prerequisites and limits
- You need segment data with positions from a platform that provides a chromosome browser, or a raw-data upload to a third-party tool.
- Short segments are often unreliable. Thresholds differ between platforms and practitioners, so check the platform documentation and use a conservative cut-off for your own work.
- Results are only as good as the underlying trees and records.
- The tool organises and visualises your data. It does not verify it, and it does not run statistical relationship tests (see the WATO and BanyanDNA guide for modelling).
Tools and further reading
🗺️ DNA Chromosome Browser
Enter segments by chromosome, start, end and length, link them to matches, tag each with a hypothesis status and add notes on sources and alternatives. Works locally on your device. It shows overlap but does not prove relationships.
📐 Guide to DNA Triangulation
The genetics behind segments, IBD and triangulated groups, and why triangulation strengthens a hypothesis without settling it.
🖥️ Intro to Chromosome Browsers
How to export segment data from MyHeritage, FamilyTreeDNA and GEDmatch, and how to spot pile-up regions.
⚖️ WATO vs. BanyanDNA
When you need probabilities rather than overlaps: comparing relationship hypotheses, and what to do when endogamy complicates the picture.
Sources for the method
- ISOGG Wiki: entries on triangulation, identical by descent and endogamy.
- The Shared cM Project (Blaine Bettinger): ranges of shared cM per relationship.
- Official help pages of MyHeritage, FamilyTreeDNA and GEDmatch for each platform's matching and threshold settings.
- Genealogical Proof Standard (Board for Certification of Genealogists) for evaluating evidence and resolving conflicts.