Thanks for trying to answer this question. I suspect there are short term variables that even out over time that interact with sex in the non coding region? For example my mom has 4 daughters and one son (me), my maternal gm has 3 daughters and one son. My materal ggm has 5 daughters and one son. I'm sure this evens out over time, but I think there is more to this than the randomness of DNA. Too bad Ancestry doesn't have a chromosome browser like gedmatch's so you would know more about your matches ancestry and be able to increase the sample size which seem to be the two biggest problems with your experiment. At about 7:58 you mention the no. 1.35. What does that refer to? You might have mentioned this in previous videos but does the 1.76 recombination number apply to all chromosomes? Specifically I'm interested in 12 right now.
A while back I saw a website or video (I can't remember which and unfortunately didn't cite it! OY) that showed the breakdown of expected segments per maternal and paternal chromosomes. It was talking about how the number of segments could possibly help to determine which side or line a match is on. So, I did the visual phasing for our three tested (adult) children, who are full siblings to each other and then created a chart to see how their segment counts worked out. For the most part, they followed the norm/average segments # expected for each chromosome, but there were 4-7 chromosomes between them that didn't match the average expected for the maternal or paternal chromosome. I'd be happy to share my chart with you (Google Sheets) if it would be any help in your research along these lines. If there's going to be an oddball in the bunch, it is going to be in a McClendon household. :) Thanks for another informative video. The math involved in these things often loses me, but I always learn something from your and Devon's videos. Thanks!
Andy, thank you for your never-ending hard work at enabling my understanding of everything dna. However, every time you unlock a new pathway of knowledge you raise more questions in my mind. If you ever come to Wales please let me know. I’ve got a million questions for you!
Thank you for answering my question regarding Neandertal variants in the last live Q & A session. Just a quick update as I know you indicated sub-Saharans do not possess Neandertal DNA, but as it turns out several groups do based on a recent Princeton peer-reviewed study. If you have time please read their paper titled, "Identifying and Interpreting Apparent Neanderthal Ancestry in African Individuals" published in the journal Cell February 20, 2020. Fascinating read!
My two cents on this… compare three first cousins … one child of an aunt, one child of an uncle and yourself… tabulate the known common relatives … but just indicate the count of genders of ancestors along pathway and then compare… for example the third cousin of yourself tMMsFMr, where t is yourself, s is a pair of ancestors, and r is your match… so in this example the 3C is M^3F… this same type of match would be grouped with your 1C from your aunt with a cousin that is tFMsMMr or even tFsMMMr… this assumes that the order of genders does not dictate the lost of cM (which is generally assumed with the avg cM per specific relationships)
And then you would compare the various pathways…. 1C: MM, MF vs FF 2C/1C2R: MMM, MMF, MFF, FFF 3C/2C2R: MMMM, MMMF, MMFF, MFFF, FFFF Etc… So a matrix representing three dimensional data can be valuable
And I’d expect the numbers to be fit to the multiplication of the recombination rates. FFF of (1.8)^3 verses MMF (1.8)*(1.3)^2 (if I remember those numbers correctly)
I think you aren't understanding average here. The sample size of 2 (you to your brother, and you to your cousin). You might just have randomly inherited more cM and more segments from your father and the common ancestor in the level above (when comparing to your cousin). You should need to be able to look at hundreds, maybe thousands of similar people to be able to see if there is anything funny going on. It would be better if you had multiple siblings and multiple cousins to be able to look at but most people don't have this luxury, and even then it might not be any different from what you've found since you might just be an outlier.
Ah, you pulled it back at the end!!! Not all maternal recombinations have this on average 1.76 times more likely to actually occur, and as I said before the sample size is too small. I think the answer to your viewer is it depends, but not always, I don't think there's a definitive answer to this question. It is more likely, but with all probability there is a range.
QUESTION: What about first-borns? I've begun to notice that I share a significant amount of more DNA with matches than my kid sister, but when she does share more with a match it's a whopping amount more. (I use multiple columns in Excel for shared cMs with Me, my sibling and parent if avail.)
Thank you for using the correct word at 1:52 (or 1:53) Hypo-thesis not theory. But though a hypothesis is less than a thesis... it is quite a lot more than an idea. Thank for not throwing in the term theory were it is incorrect to use it :) But what is that damn theory you mention at 10:05? As for what is going on at your "measurement experiment" you have too small a sample size. You are looking at percentages and those do not really work in a sample size of what "2" ? you need every relative who has taken a DNA test and every one needs to be related to every one... That is the big Failure there! n =100 not n=1
At 0:44, you state that "We know maternal chromosomes have 1.76 times the number of recombinations as paternal chromosomes." Really? Is this for homo sapiens? For a single population or the average for ALL populations? What causes this sex-linked phenomenon in autosomal chromosomes? And what is your source for this "fact"??
Thanks for trying to answer this question. I suspect there are short term variables that even out over time that interact with sex in the non coding region? For example my mom has 4 daughters and one son (me), my maternal gm has 3 daughters and one son. My materal ggm has 5 daughters and one son. I'm sure this evens out over time, but I think there is more to this than the randomness of DNA. Too bad Ancestry doesn't have a chromosome browser like gedmatch's so you would know more about your matches ancestry and be able to increase the sample size which seem to be the two biggest problems with your experiment. At about 7:58 you mention the no. 1.35. What does that refer to? You might have mentioned this in previous videos but does the 1.76 recombination number apply to all chromosomes? Specifically I'm interested in 12 right now.
A while back I saw a website or video (I can't remember which and unfortunately didn't cite it! OY) that showed the breakdown of expected segments per maternal and paternal chromosomes. It was talking about how the number of segments could possibly help to determine which side or line a match is on. So, I did the visual phasing for our three tested (adult) children, who are full siblings to each other and then created a chart to see how their segment counts worked out. For the most part, they followed the norm/average segments # expected for each chromosome, but there were 4-7 chromosomes between them that didn't match the average expected for the maternal or paternal chromosome.
I'd be happy to share my chart with you (Google Sheets) if it would be any help in your research along these lines. If there's going to be an oddball in the bunch, it is going to be in a McClendon household. :)
Thanks for another informative video. The math involved in these things often loses me, but I always learn something from your and Devon's videos. Thanks!
Andy, thank you for your never-ending hard work at enabling my understanding of everything dna. However, every time you unlock a new pathway of knowledge you raise more questions in my mind. If you ever come to Wales please let me know. I’ve got a million questions for you!
Thank you for answering my question regarding Neandertal variants in the last live Q & A session. Just a quick update as I know you indicated sub-Saharans do not possess Neandertal DNA, but as it turns out several groups do based on a recent Princeton peer-reviewed study. If you have time please read their paper titled, "Identifying and Interpreting Apparent Neanderthal Ancestry in African Individuals" published in the journal Cell February 20, 2020. Fascinating read!
I'm glad that Andy was able to address your concerns about Neanderthal DNA.
My two cents on this… compare three first cousins … one child of an aunt, one child of an uncle and yourself… tabulate the known common relatives … but just indicate the count of genders of ancestors along pathway and then compare… for example the third cousin of yourself tMMsFMr, where t is yourself, s is a pair of ancestors, and r is your match… so in this example the 3C is M^3F… this same type of match would be grouped with your 1C from your aunt with a cousin that is tFMsMMr or even tFsMMMr… this assumes that the order of genders does not dictate the lost of cM (which is generally assumed with the avg cM per specific relationships)
And then you would compare the various pathways….
1C: MM, MF vs FF
2C/1C2R: MMM, MMF, MFF, FFF
3C/2C2R: MMMM, MMMF, MMFF, MFFF, FFFF
Etc…
So a matrix representing three dimensional data can be valuable
And I’m sorry that I can’t tabulate this myself… not enough 1C… plus endogamy on paternal side… and most maternal are 3C and futher
And I’d expect the numbers to be fit to the multiplication of the recombination rates. FFF of (1.8)^3 verses MMF (1.8)*(1.3)^2 (if I remember those numbers correctly)
I think you aren't understanding average here. The sample size of 2 (you to your brother, and you to your cousin). You might just have randomly inherited more cM and more segments from your father and the common ancestor in the level above (when comparing to your cousin). You should need to be able to look at hundreds, maybe thousands of similar people to be able to see if there is anything funny going on. It would be better if you had multiple siblings and multiple cousins to be able to look at but most people don't have this luxury, and even then it might not be any different from what you've found since you might just be an outlier.
Ah, you pulled it back at the end!!! Not all maternal recombinations have this on average 1.76 times more likely to actually occur, and as I said before the sample size is too small. I think the answer to your viewer is it depends, but not always, I don't think there's a definitive answer to this question. It is more likely, but with all probability there is a range.
QUESTION: What about first-borns? I've begun to notice that I share a significant amount of more DNA with matches than my kid sister, but when she does share more with a match it's a whopping amount more. (I use multiple columns in Excel for shared cMs with Me, my sibling and parent if avail.)
Great question!
Glad you think so.
Thank you for using the correct word at 1:52 (or 1:53) Hypo-thesis not theory. But though a hypothesis is less than a thesis... it is quite a lot more than an idea. Thank for not throwing in the term theory were it is incorrect to use it :)
But what is that damn theory you mention at 10:05?
As for what is going on at your "measurement experiment" you have too small a sample size. You are looking at percentages and those do not really work in a sample size of what "2" ? you need every relative who has taken a DNA test and every one needs to be related to every one... That is the big Failure there! n =100 not n=1
At 0:44, you state that "We know maternal chromosomes have 1.76 times the number of recombinations as paternal chromosomes." Really? Is this for homo sapiens? For a single population or the average for ALL populations? What causes this sex-linked phenomenon in autosomal chromosomes? And what is your source for this "fact"??
I have less matches on my mothers side. 🧐😢
My wife has fewer on her father's. Welcome to family dynamics.