Frequently Asked Questions

Before you start you'll need:

• The name and birth year of someone with an unknown parent
• The name of the person whose DNA matches you’re using (and their relationship to the person with the unknown parent (e.g. child, grandchild)
• A family tree that connects DNA matches

The first step would be to go to the detail page for your top matches and then look at shared matches. You can then organize your matches into clusters (also known as genetic networks). Both Ancestry and MyHeritage have a system with coloured dots that can simplify this process, but you might prefer just using the Leeds Method with Excel or paper and pencils.

If you are able to, you can then try to figure out which cluster or clusters are relevant to the unknown parent. Your task is then to use whatever information these DNA matches have provided and try to work out their genealogical connection to each other. Once you've done this, you'll be ready to use WATO plus.

Yes, in order to use the tool you need a free or paid DNA Painter account so that your tree can be saved securely.

No, you will need a subscription for full access to WATO plus. Anyone can create a WATO plus tree but a subscription is required to create additional trees. Functionality within WATO plus trees will also in future be limited slightly for non-subscribers, with a subscription required to import a text file or access age-based probabilities.

Subscriptions are currently frozen at their 2018 price of $55 for a year, with subscribers able to create up to 50 WATO plus trees alongside other subscription benefits in the ancestral trees and chromosome mapping applications.

The requirement to subscribe for full access to WATO plus will help fund ongoing development and refinement as well as all previous work to date.

The main differences are as follows:

• The tool asks for more specific information up front and is therefore able to make more useful suggestions.
  • For example, while in the old tool you just stated the birth year of the DNA tester, in the new version, you give the birth year of the person who has the unknown parent (this could be for example the parent, grandparent or other ancestor of the tester)
  • If for example you state you're looking for the unknown father of your mother Kathy, whenever you create a hypothesis, WATO plus will create a node for Kathy and a node for you.
  • The tool will also know only to suggest males as candidates to be the unknown father
• The Hypothesis is now the person you think was the unknown parent
  • In previous versions, you added a hypothesis to each place where you thought the DNA tester might fit into the tree. This worked fine in unknown parentage cases, but got confusing for other scenarios such as unknown grandparents
• Scores are now given as percentages
  • In previous versions, the score was a number. More info below under scores.
• You can now take account of the age of the parent
  • More info below under probabilities.

While this new version has undergone extensive testing, there may still be bugs. I've therefore kept a BETA label on WATO plus. I hope to develop and refine it more in response to feedback from the genealogy community.

Having figured out how two or more DNA matches connect to each other, you can get this information into WATO plus as follows:

• Import a GEDCOM file
  • Just as in the older versions, you can browse for a GEDCOM file (which can be exported from most family tree websites and software).
  • You can find this when you click actions and settings
  • You would then choose the couple whose descendants you want to import
• Import a text file from WATO plus or a previous version of WATO
  • You can download any tree in text format and then import it via Import text file in the actions and settings menu
• You can also create the tree manually by simply hovering over the starting node Most recent common ancestor or couple and clicking add child and then repeating the process as necessary.
• If you choose to build your tree this way, please be sure to hover and click Edit details and add a birth year for as many people as possible. This will help WATO plus make better suggestions

This works exactly as in previous WATO versions: simply hover over the person who is a DNA match and click Enter match cM

If you've imported a text file, the matches will already be marked.

There are two ways to tell WATO plus that you want to explore the hypothesis that a person was the unknown parent:

• Manually: hover over the person and click use as hypothesis
• Automatic: click the generate hypotheses button

The amount that the child shares with the tester is not relevant to the overall calculation, so this is just the regular message that WATO outputs when cMs for a child are entered. These ‘child’ amounts are not used.

For example, imagine you have a match called David, sharing 200cM, who is a possible 2nd cousin.

• David might have a child, and he might pass down 90cM of those 200cM down to this child.
• But the fact that the child inherited 90cM as opposed to say 40cM or 180cM does not help you figure out whether or not David is your 2nd cousin, since all the DNA you share with the child is just a subset of what you share with David.
• Therefore if you enter the cMs you share with David's child as well as those you share with David, you'll see the 'Parent cM used' note above the child's node

WATO originally used probabilities simulated by AncestryDNA and originally described in their 2016 White Paper.

In 2020, updated probabilities were compiled from probabilities presented at ancestry.com/dna, as documented by Leah Larkin. These 2020 probabilities are currently the default, with the option to switch to the 2016 White Paper probabilities.

Additional probability options will be added in future.

Both sets of probabilities have strengths and weaknesses. I'm hoping to have improved probabilities in future, but in the meantime:

• The default 2020 set of probabilities handle close relationships well, reflecting for example the broad range of amounts that a first cousin once removed (1C1R) can share.
  • However, for low cM amounts, the probabilities seem to favour close relationships far more strongly than they should. While there isn't a consensus, it is widely believed that even a segment of 20cM can be as old as 10 to 20 generations, yet the most probable relationships given are Half 3C, 3C1R, Half 2C2R and 2C3R. More distant relationships are given a much lower probability
  • In practical terms, this means that if you have mostly matches under 40cM, those hypotheses that suggest more distant relationships like 4C1R or more distant may have very low scores due to low probabilities being compounded.
• The 2016 probabilities have what based on my experience is a more sensible set of probabilities for smaller numbers of cMs shared. For example, 20cM of shared DNA is given a 62% probability of being a 4C1R or more distant.
  • However, for closer relationships like 1C1R, the permitted ranges are a bit tighter, making these probabilities less useful for the most "solvable" cases.

If a tree has three or more matches of under 50cM and the default 2020 probabilities are being used, a message will appear suggesting that the user switch to the 2016 probabilities for more accurate results.

Click Options and Settings and you'll see the option under Probabilities.

As soon as you click an option here, it will take effect immediately. Your browser will then default to this set of probabilities in future

When considering whether someone in the tree could have been the parent, the tool can consider their age at the time the child was born.

This feature can be useful if you have a number of potential candidates born in different years. It works by estimating the probability that the mother or father of a child born that year was a certain age. For example, you might be looking for the father of David born in 1945, and you might have candidates Bill and Eric. When the box is checked, it will calculate how old Bill and Eric were in 1945 and then apply a multiplier based on actual birth data for that year.

Please note: if you have hypotheses that are based on 'Unknown' theoretical people that WATO plus has added when you clicked Suggest Hypotheses, then this option will be less useful since those ages will be estimates for people who may or may not have existed.

The data currently used is from various editions of the US National Vital Statistics Report from 1936-2014.

Additional likelihood data from other countries will be added in future.

Below your research question at the top left, you'l see two checkboxes:

• Apply [parent]'s age to the probabilities
• Also apply [other biological parent]'s age to the probabilities

Please note: in order for this box to become active, you need to indicate the specific parent (mother or father) that you are looking for. This is because the probability data is sex-specific).

The age-based adjustments will be made as soon as you check the box, and you can see the details in the Scores table at the bottom. You can reach this by clicking Scores in the fixed menu on the right of the screen.

WATO plus will take everyone in the tree and test out the hypothesis that they were the missing parent.

• Depending on your settings, the tool will also create new nodes for people who don't exist in the tree in places where the DNA suggests they could be a feasible candidate.
• The tool can also optionally create additional sibling and ancestor nodes to the root node in the tree.

You can read a blog post explaining this in more detail.

When suggesting hypotheses, if you have Stop WATO from adding unknown people to the tree unchecked, WATO plus will attempt to add a node in every location where it would create a distinct score.

• So if for example Larry has a tested descendant but no full sibling in the tree, it would create one for him
• But if Larry already has a full sibling with no tested descendants, WATO plus will just use that full sibling as a hypothesis and it won't create a new person since they would have the same score.

If you are confident you have all the possible candidates in the tree, you could check Stop WATO from adding unknown people to the tree.

If you're confident that the DNA tester is related to both of the common ancestral couple, you can keep Allow WATO to add siblings of the root person turned off.

You can read a blog post explaining this in more detail.

If you have no suggestions, this suggests that either:

• The tree is not correct, e.g. someone is in the wrong generation.
• There is an error in the cM amounts entered for one or more matches
• There is pedigree collapse or endogamy in the tree. Both of these can inflate the amounts of DNA shared.

Within WATO plus, you can use the Flex option within Actions and Settings.

• This will automatically reduce the entered cM amounts by between 5 and 50% and use these figures as the basis for probabilities.
• If you adjust this setting and then click Suggest hypotheses you may get some suggestions
• You can then move the Flex setting back to zero and take a look at which matches are invalidating these hypotheses