FST - התשובה של פרופ' גרייבס
בדיון האחרון עלתה שאלה לגבי מדד FST לקביעת גזעים ובמיוחד לגבי בני אדם מודרניים.
לחבר הפורום צ'ארלס היתה השגה לגבי למה חוקר כמו גרייבס משתמש בכלל ב FST והאם הוא משתמש במדד בהתאם לכוונה של סיוול רייט.
רק-עובר טען שהשימוש המקורי במדד נחשב למיושן ויש מחלוקת, גם עפ"י רייט עצמו, לגבי האפליקטיביות שלו בעולם האמיתי.
הפניתי את השאלה לפרופ' גרייבס וכפי ששערתי הוא הואיל בטובו לענות תשובה מפורטת.
אשמח לשמוע את השגות מומחיינו בנושא (גרייבס קלט שאני בור גדול בגנטיקה של אוכלוסיות) ואנסה להעביר לגרייבס אם יהיה טעם.
In answer to your question below. The Hartl and Clark description of Fst relation to genetic variation is correct. Thus, when we are talking about the biological definition of race we are concerned with how much variation exists between populations that we suspect are geographic races. This is true of all species (not just humans.) Thus, when Sewall Wright proposed in 1968 that the lowest threshold for the existence of a biological or geographic race as Fst = 0.250, he was saying that we describe populations that have very great genetic differentiation as biological or geographic races.
The socially-defined groups within our species (anatomically modern humans) do not have any where near that much differentiation averaged across the genome. This is why I categorically state that our species does not have biological races. Socially-defined races differ by historical period and society. For example in Brazil, there are at least 12 categories of persons who have African descent. These categories are real and meaningful within Brazilian society, so a "moreno" differentiates themselves from a "clara." In the United States both groups would be called "black" or "African American."
While I have stated above that the average Fst for the human genome does not approach 0.250, there are specific loci which do exceed the 0.250 figure. Anything which is subject to strong natural selection associated with a geographic cline (such as sunlight) is likely to have a strong Fst value. Skin color associated loci have large Fst values, but remember there are only five (of 25,000 coding loci) that code for skin color. This is why geneticists like Rick Kittles (U. Chicago) and Ken Shriver (Penn. St.) can utilize specific genetic markers to estimate the amount of continental ancestry that individuals have. These are called ancestry informative markers (AIMs), and they have been used to estimate the percentage of European genes in African Americans (~ 16% by region.) The percentage of AIMs in the entire human genome are less than 5% of its composition.
You are correct that intelligence is a complex trait with very many genetic loci contributing to it. I have pointed our for a long time that intelligence is not racialized. I have chapters on this in both of my books, The Emperor's New Clothes: Biological Theories of Race at the Millennium, Rutgers U. Press 2005 and The Race Myth: Why We Pretend Race Exists in America, Dutton Books, 2005. I just wrote a piece on the genomics of intelligence that will be coming out later this year from Nova Press. You incorrect in asserting that one cannot study the Fst associated with genetic variation that has been associated with intelligence. A large number of genetic markers have been related to high intellectual performance and each of these has a very large number of single nucleotide polymorphisms (SNPs) within them. In the Nova chapter I analyzed the Fst for these genetic variants and found that 95% had Fst below 0.250. In other words genes associated with cognitive function have Fst distributions just like the rest of the human genome.
In summary, I strongly recommend that you actually read my books on biological conceptions of race (as opposed to snippets taken from Utube.) I think you misunderstood my views. In addition, it would profit you to take a course in evolutionary or population genetics, because you cannot understand the genetic/genomic issues associated with race, without that type of training.
Sincerely,
Dr. Joseph L. Graves Jr.
Assoc. Dean for Research
JSNN
Greensboro, NC 27401
שאלת ההדיוט:
Hello Prof. Graves,
In one of your talks, Spelman College Oct 2012, you mentioned that “Sewall Wright felt that Fst = 0.25 was required to define geographic races and subspecies”:
http://www.youtube.com/watch?feature=player_detailpage&v=cF4VxSA3zVQ#t=1463s
After some web research I found the following definition that is attributed to Wright (Hartl and Clark 1989. p118-119):
The range 0 to 0.05 may be considered as indicating little genetic differentiation.
The range 0.05 to 0.15 indicates moderate genetic differentiation.
The range 0.15 to 0.25 indicates great genetic differentiation.
Values of FST above 0.25 indicate very great genetic differentiation.
Is this consistent with the 0.25 claim that you have mentioned during the talk?
Why FST is actually important for defining human races for which the only phenotype that matters is the cognitive one for which, as far as I can tell, we don’t know the FST?
For example, I saw claim that the FST for skin color has been estimated to be about 0.6 (Relethford, J.H. 1992) yet no one is going use this trait to define races.
I hope that you will be able to find time to address my questions.
Thanks,
Rani
בדיון האחרון עלתה שאלה לגבי מדד FST לקביעת גזעים ובמיוחד לגבי בני אדם מודרניים.
לחבר הפורום צ'ארלס היתה השגה לגבי למה חוקר כמו גרייבס משתמש בכלל ב FST והאם הוא משתמש במדד בהתאם לכוונה של סיוול רייט.
רק-עובר טען שהשימוש המקורי במדד נחשב למיושן ויש מחלוקת, גם עפ"י רייט עצמו, לגבי האפליקטיביות שלו בעולם האמיתי.
הפניתי את השאלה לפרופ' גרייבס וכפי ששערתי הוא הואיל בטובו לענות תשובה מפורטת.
אשמח לשמוע את השגות מומחיינו בנושא (גרייבס קלט שאני בור גדול בגנטיקה של אוכלוסיות) ואנסה להעביר לגרייבס אם יהיה טעם.
In answer to your question below. The Hartl and Clark description of Fst relation to genetic variation is correct. Thus, when we are talking about the biological definition of race we are concerned with how much variation exists between populations that we suspect are geographic races. This is true of all species (not just humans.) Thus, when Sewall Wright proposed in 1968 that the lowest threshold for the existence of a biological or geographic race as Fst = 0.250, he was saying that we describe populations that have very great genetic differentiation as biological or geographic races.
The socially-defined groups within our species (anatomically modern humans) do not have any where near that much differentiation averaged across the genome. This is why I categorically state that our species does not have biological races. Socially-defined races differ by historical period and society. For example in Brazil, there are at least 12 categories of persons who have African descent. These categories are real and meaningful within Brazilian society, so a "moreno" differentiates themselves from a "clara." In the United States both groups would be called "black" or "African American."
While I have stated above that the average Fst for the human genome does not approach 0.250, there are specific loci which do exceed the 0.250 figure. Anything which is subject to strong natural selection associated with a geographic cline (such as sunlight) is likely to have a strong Fst value. Skin color associated loci have large Fst values, but remember there are only five (of 25,000 coding loci) that code for skin color. This is why geneticists like Rick Kittles (U. Chicago) and Ken Shriver (Penn. St.) can utilize specific genetic markers to estimate the amount of continental ancestry that individuals have. These are called ancestry informative markers (AIMs), and they have been used to estimate the percentage of European genes in African Americans (~ 16% by region.) The percentage of AIMs in the entire human genome are less than 5% of its composition.
You are correct that intelligence is a complex trait with very many genetic loci contributing to it. I have pointed our for a long time that intelligence is not racialized. I have chapters on this in both of my books, The Emperor's New Clothes: Biological Theories of Race at the Millennium, Rutgers U. Press 2005 and The Race Myth: Why We Pretend Race Exists in America, Dutton Books, 2005. I just wrote a piece on the genomics of intelligence that will be coming out later this year from Nova Press. You incorrect in asserting that one cannot study the Fst associated with genetic variation that has been associated with intelligence. A large number of genetic markers have been related to high intellectual performance and each of these has a very large number of single nucleotide polymorphisms (SNPs) within them. In the Nova chapter I analyzed the Fst for these genetic variants and found that 95% had Fst below 0.250. In other words genes associated with cognitive function have Fst distributions just like the rest of the human genome.
In summary, I strongly recommend that you actually read my books on biological conceptions of race (as opposed to snippets taken from Utube.) I think you misunderstood my views. In addition, it would profit you to take a course in evolutionary or population genetics, because you cannot understand the genetic/genomic issues associated with race, without that type of training.
Sincerely,
Dr. Joseph L. Graves Jr.
Assoc. Dean for Research
JSNN
Greensboro, NC 27401
שאלת ההדיוט:
Hello Prof. Graves,
In one of your talks, Spelman College Oct 2012, you mentioned that “Sewall Wright felt that Fst = 0.25 was required to define geographic races and subspecies”:
http://www.youtube.com/watch?feature=player_detailpage&v=cF4VxSA3zVQ#t=1463s
After some web research I found the following definition that is attributed to Wright (Hartl and Clark 1989. p118-119):
The range 0 to 0.05 may be considered as indicating little genetic differentiation.
The range 0.05 to 0.15 indicates moderate genetic differentiation.
The range 0.15 to 0.25 indicates great genetic differentiation.
Values of FST above 0.25 indicate very great genetic differentiation.
Is this consistent with the 0.25 claim that you have mentioned during the talk?
Why FST is actually important for defining human races for which the only phenotype that matters is the cognitive one for which, as far as I can tell, we don’t know the FST?
For example, I saw claim that the FST for skin color has been estimated to be about 0.6 (Relethford, J.H. 1992) yet no one is going use this trait to define races.
I hope that you will be able to find time to address my questions.
Thanks,
Rani
