Генеалогические построения по результатам от Relative Finder

[I2a1a]

Yes, the "cluster tree" was built using the whole set of regions of 6th chromosome.

[I2a1a]

Добавил еще 6 человек и построил два дерева - одно по методу Minimum Evolution, другое по методу Фич-Марголиаша.

[warwick]

Добавил еще 6 человек и построил два дерева - одно по методу Minimum Evolution, другое по методу Фич-Марголиаша.

Given the clear differences, is there a reason to think that one tree is more accurate or is that dependent on different assumptions. Very interesting Vadim.

[I2a1a]

Добавил еще 6 человек и построил два дерева - одно по методу Minimum Evolution, другое по методу Фич-Марголиаша.

Given the clear differences, is there a reason to think that one tree is more accurate or is that dependent on different assumptions. Very interesting Vadim.

Assuming that both trees have been obtained from the same "genetic distance" matrix, one can not say for sure, which method -Neighbor-Joining or Fitch Square Distances - will give more accurate (plausible in terms of genealogy) results. Perhaps neither of these methods is applicable to the clustering of individuals according to their matrix relationships (because there are only "genealogical distances" without supporting molecular data) but otherwise -withouth testing both methods on this matrix -we would have never known that.

Nevertheless, i've spent a few hours playing with ATV applet
http://phylogeny.lirmm.fr/phylo_cgi/one_task.cgi?task_type=atv
I've been able to produce somehow more meaningful cluster scheme

[I2a1a]

Later that week i'll be going to add these folks to my cluster scheme.

Vadim   Verenich   Alex   Brennen   6   4   2666   24917750   32519502   7601752   11
Vadim   Verenich   Sean   MacGorman   6   3.3   2505   25411079   32489918   7078839   14
Vadim   Verenich   Mary   Powell   6   3.2   411   166186209   167346672   1160463   21
Vadim   Verenich   Paul   Oksnee   6   3.1   507   14444452   16528289   2083837   19
Vadim   Verenich   Lesley   Murphy   6   3.1   483   153543401   155307348   1763947   20
Vadim   Verenich   James   Lick   6   2.9   2426   26433867   32906526   6472659   16
Vadim   Verenich   Lesley   Uncle   6   2.6   411   123512254   125016923   1504669   24
Vadim   Verenich   Robert   Powell   6   2.5   555   152566535   154464438   1897903   24
Vadim   Verenich   Gary   Corbett   6   2.5   616   73595193   77374403   3779210   21
Vadim   Verenich   Sean   MacGorman   6   2.4   452   20317323   22067417   1750094   25
Vadim   Verenich   James   Lick   6   2.4   450   125410408   128614518   3204110   22
Vadim   Verenich   Robert   Powell   6   2.3   487   151269402   152479797   1210395   29
Vadim   Verenich   Lesley   Uncle   6   2.3   416   21798151   23283063   1484912   28
Vadim   Verenich   Gary   Corbett   6   2.2   549   121077353   123944241   2866888   26
Vadim   Verenich   James   Lick   6   2.2   413   142624234   144415612   1791378   28
Vadim   Verenich   Sean   MacGorman   6   2.1   424   151480681   152563916   1083235   33
Vadim   Verenich   Lesley   Murphy   6   2.1   408   87016654   89230478   2213824   29

Then the true fun will begin!

[warwick]

Later that week i'll be going to add these folks to my cluster scheme.

Vadim   Verenich   Alex   Brennen   6   4   2666   24917750   32519502   7601752   11
Vadim   Verenich   Sean   MacGorman   6   3.3   2505   25411079   32489918   7078839   14
Vadim   Verenich   Mary   Powell   6   3.2   411   166186209   167346672   1160463   21
Vadim   Verenich   Paul   Oksnee   6   3.1   507   14444452   16528289   2083837   19
Vadim   Verenich   Lesley   Murphy   6   3.1   483   153543401   155307348   1763947   20
Vadim   Verenich   James   Lick   6   2.9   2426   26433867   32906526   6472659   16
Vadim   Verenich   Lesley   Uncle   6   2.6   411   123512254   125016923   1504669   24
Vadim   Verenich   Robert   Powell   6   2.5   555   152566535   154464438   1897903   24
Vadim   Verenich   Gary   Corbett   6   2.5   616   73595193   77374403   3779210   21
Vadim   Verenich   Sean   MacGorman   6   2.4   452   20317323   22067417   1750094   25
Vadim   Verenich   James   Lick   6   2.4   450   125410408   128614518   3204110   22
Vadim   Verenich   Robert   Powell   6   2.3   487   151269402   152479797   1210395   29
Vadim   Verenich   Lesley   Uncle   6   2.3   416   21798151   23283063   1484912   28
Vadim   Verenich   Gary   Corbett   6   2.2   549   121077353   123944241   2866888   26
Vadim   Verenich   James   Lick   6   2.2   413   142624234   144415612   1791378   28
Vadim   Verenich   Sean   MacGorman   6   2.1   424   151480681   152563916   1083235   33
Vadim   Verenich   Lesley   Murphy   6   2.1   408   87016654   89230478   2213824   29

Then the true fun will begin!

How much time does it take to generate these cluster trees for these chr 6 datasets? Is there a limit to the number of data points that can be analyzed in a reasonable time frame, or does the software run very fast?

[I2a1a]

The generation of these cluster trees in itself doesn't take a lot of time. It's the calculation of distance matrices that is very time consuming, because there is no software to calculate distance for HIR matches, i was calculating them manually.

[napobo3]

Guys, I enjoy your discussion, but some forum participants don't feel comfortable with a foreign language. I don't mean to tell you "go to dna-forums to discuss this". We need to find a solution.

[warwick]

Guys, I enjoy your discussion, but some forum participants don't feel comfortable with a foreign language. I don't mean to tell you "go to dna-forums to discuss this". We need to find a solution.

I'm just using google translate. Vadim may or may not want discuss this on dna-forums. Isn't google translate english-russian accessible to the other members?
I do not have any Russian language knowledge myself, but google translate seems to be effective (thus far...)

[napobo3]

Guys, I enjoy your discussion, but some forum participants don't feel comfortable with a foreign language. I don't mean to tell you "go to dna-forums to discuss this". We need to find a solution.

I'm just using google translate. Vadim may or may not want discuss this on dna-forums. Isn't google translate english-russian accessible to the other members?
I do not have any Russian language knowledge myself, but google translate seems to be effective (thus far...)

Fine. That's the solution. Justin will use Google translate and Vadim will reply in Russian.
We cannot require from other members to use Google translate because they are participating in the forum that declares itself as a forum-in-Russian-language.

[warwick]

Guys, I enjoy your discussion, but some forum participants don't feel comfortable with a foreign language. I don't mean to tell you "go to dna-forums to discuss this". We need to find a solution.

I'm just using google translate. Vadim may or may not want discuss this on dna-forums. Isn't google translate english-russian accessible to the other members?
I do not have any Russian language knowledge myself, but google translate seems to be effective (thus far...)

Fine. That's the solution. Justin will use Google translate and Vadim will reply in Russian.
We cannot require from other members to use Google translate because they are participating in the forum that declares itself as a forum-in-Russian-language.

Although google translate is good for russian -> english, I agree that responding in English is problematic.

[I2a1a]

Ок.

[I2a1a]

Добавил Sean McGorman Powell и Alex Brennen.

Кластеры начинают вырисовыватся более точно

[I2a1a]

В идеале - должно получится что-то вроде этого.

Привязка беларуса к эльзаско-швейцарско-австрийскому кластеру уже имеется.

[napobo3]

Добавил Sean McGorman Powell и Alex Brennen.

Кластеры начинают вырисовыватся более точно

Вадим, выбор моего двоюродного дяди (Mike Druyan) был не очень удачным на первом этапе отладки алгоритмов. 50% ашкеназ, 50% русский, плюс протестированный не в 23&ми, а в ФДТНА, поэтому результаты сравнения в 4 раза грубее. Он не должен находиться в отчетах упсометра, пока мы не запустим мультичип-сравнения, так уж вышло, что он там.