The J2 Y-DNA project respects the wishes of its members with regard to whether their haplotype is in the public domain or not. Only the haplotypes of project members who consent are displayed on this page. Any member who does not wish their haplotype to be in the public domain will not be obligated to do so.
**IMPORTANT**
These results are arranged according to the haplotype cluster that the project member is in, or is likely to be in.
(see the project phase three analysis pages for more information on the defined clusters).
The clusters are defined by at least 37 markers. If the haplotype has 25 markers or less, In some instances we can predict what cluster the haplotype might be in, but in other instances we cannot. For haplotypes that only have 12 markers we usually cannot reliably predict what cluster a haplotype might be in (beyond whether it is likely to be J2a (M410+) or J2b (M12+, M102+). In some instances we cannot even reliably predict whether a 12 marker haplotype is in J2a or J2b. See here for additional information regarding 12 marker haplotypes.
Bear in mind, if you have less than 37 markers and we predict which cluster your haplotype might be in, we cannot guarantee 100% beyond a doubt that the haplotype is indeed in that cluster. Generally though, we have been scientifically conservative in making our predictions (ie. we prefer to present the predictions only if the probability is close to the 95% probability level that is used in science; other less scientific people might provide predictions with only a 50% level of probability (yet give the impression that the predictions are 100% definitive).
If you have less than 37 markers, and would like to know what cluster you are likely to be in with more certainty, then we would advise you to upgrade to at least 37 markers. With 37 markers, most members will be placed within a cluster, however there will still be a minority of members that will still be unplaced (usually due to having an unusual haplotype and/or not clustering closely with others in the project).
What do we mean by "reliably predict" ?
What we mean by "reliably predict", is predict with at least 95% accuracy (preferably 99-100% accuracy).
This accuracy threshold of 95% used by the project corresponds to the statistical probability of 95% that is used for accepting scientific hypotheses (ie. we are following standard scientific procedure).
If we can predict with 95-100% accuracy, we will present the predictions. Bear in mind that the practical reality of "95% statistical certainty" means that out of 100 predictions, 5 will be wrong.
If there are 100 members with a prediction of 95% probability, this means that:
• 95 members will have a correct prediction, ..... however it also means that :
• 5 members will have an incorrect prediction.
If you receive a prediction of 95% probability, it might be tempting to regard the prediction as fact, but bear in mind it is possible that you are part of the 5% for which the prediction is wrong.
For the 12 marker haplotypes we have been extra conservative. More information about this can be seen here.
For some haplotypes we can predict at levels of 50-80 % accuracy, but we generally don't present these predictions (since it is below the probability level used for accepting scientific hypotheses).
The project will actively revise its predictions, based on new haplotype data and/or project analyses (eg. updated network diagrams, future updates to the phase 2 and phase 3 analyses).
Extra Information about cluster assignments for members that have only 12 markers.
If a haplotype has only 12 markers, it is very difficult to reliably assign it to a cluster. This is because the first 12 markers are on average, of low diagnostic value (see here for additional information). More information will be given at the next analysis update.
- Therefore, we cannot assume that matching 12 marker haplotypes are from the same clade and/or cluster. For this reason we cannot reliably assign 12 marker haplotypes to a cluster
