Crimson Guard

May 29 2010, 04:25 AM

I'm not too interested in genetics to be honest, so cant be of much help, CB. I do know familytree is behind in the times and not very reliable on the origins of the markers.

mtDNA K is certainly not Jewish:

http://en.wikipedia.org/wiki/Haplogroup_K_%28mtDNA%29

E3b seems to be a Caucasoid( or proto-Caucasoid) marker, one sub-type or clade(whatever you want to call it) is mostly limited to the Berbers and another clade is indigenous Southeastern Europe/Balkans which is the one the bulk of what Europeans have.

Odin of Ossetia

May 29 2010, 10:00 PM

You guys are retarded.


It is universally known that the J paternal lines are of Semitic origin.

Curious2know

May 30 2010, 12:25 AM

It was the Tainos who had "recently" arrived in Cuba. Most Cuban natives were Siboney and they had been in Cuba for thousands of years. However, this group's ancestry encompassed those who had migrated from Fla. as well as the Yucatan and South America, according to the info. I have read. I think that Cuba DNA Project is a good representation of Cuban-Americans Y and mitochondrial DNA, it is not a autosomal test, after all.

Attached to this post:

cu12.jpg (72.08 KB)

cu20__Small_.jpg (12.15 KB)

cuban1xi8.jpg (56.65 KB)

cuban2tt9.jpg (58.84 KB)

cu10__Medium_.jpg (116.6 KB)

cuban6jf6.jpg (99.09 KB)

cuban15kf9.jpg (63.34 KB)

Crimson Guard

May 30 2010, 06:37 AM

Cuban-Basque

May 30 2010, 12:34 AM

but still it shows that 42% of MtDNA is Euro, being the biggest group of all, and 70% of Y-DNA is Euro also.

This is an older one, but based on all the genetics, how well does this figure for Cuba correlate you think?



http://s1.zetaboards.com/anthroscape/topic/3139425/1/#new

Curious2know

May 30 2010, 05:14 PM

Cuban-Basque

May 30 2010, 03:09 AM

The amount of mestizos that married into the mainstream white Cuban population was very little, so that is the exception not the rule

I think you are adhering to a dogma and not reality. DNA tells you what really happened during the first two centuries of colonization. The fact that so many Cubans have Native American mitochondrial hgs tells you a lot. The fact that so many have European mtDNA, compared to others in the region, tells you a lot of Spanish and other Euro women settled in Cuba and that whole families became part of the colonization process. The early mestizo population was simply absorbed or incorporated into the Spanish population. Apparently, there never developed a separate mestizo population in Cuba. And this classification, meaning people of mixed Euro-Native American ancestry, was not used or simply discarded. At one point, there were no longer any identifiable natives and the amount of "Indian" blood was minor. Having a Native American or sub-Saharan hg, is not an indictation of someone's actual genetic make up. Only an autosomal test will tell you that. I would bet anything that those Cubans who joined the Cuba DNA Project were very surprised to find out they had Amerindian hgs, on the maternal side. And most who have a sub-Saharan hg probably did not expect it either.

Quote:

 

Abstract
Background
Before the arrival of Europeans to Cuba, the island was inhabited by two Native American groups, the Tainos and the Ciboneys. Most of the present archaeological, linguistic and ancient DNA evidence indicates a South American origin for these populations. In colonial times, Cuban Native American people were replaced by European settlers and slaves from Africa. It is still unknown however, to what extent their genetic pool intermingled with and was 'diluted' by the arrival of newcomers. In order to investigate the demographic processes that gave rise to the current Cuban population, we analyzed the hypervariable region I (HVS-I) and five single nucleotide polymorphisms (SNPs) in the mitochondrial DNA (mtDNA) coding region in 245 individuals, and 40 Y-chromosome SNPs in 132 male individuals.
Results
The Native American contribution to present-day Cubans accounted for 33% of the maternal lineages, whereas Africa and Eurasia contributed 45% and 22% of the lineages, respectively. This Native American substrate in Cuba cannot be traced back to a single origin within the American continent, as previously suggested by ancient DNA analyses. Strikingly, no Native American lineages were found for the Y-chromosome, for which the Eurasian and African contributions were around 80% and 20%, respectively.
Conclusion
While the ancestral Native American substrate is still appreciable in the maternal lineages, the extensive process of population admixture in Cuba has left no trace of the paternal Native American lineages, mirroring the strong sexual bias in the admixture processes taking place during colonial times.

Quote:

 

Samples
Blood stains from 245 unrelated individuals from the general Cuban population were collected blindly with regard to their socioeconomic status in order to avoid ascertainment biases. For each individual, information on the province of origin of the maternal grandmother and paternal grandfather was recorded (see Figure ​Figure1)1) for further analyses on the geographical distribution of the lineages within Cuba. Samples consisted of unrelated healthy blood donors and appropriate informed consent was obtained from all individuals participating in the study. DNA was extracted using standard phenol-chloroform protocols.

Quote:

 

Phylogeographic analysis of mtDNA lineages in Cuba
A total of 153 different mtDNA sequences were found in the 245 individuals analyzed. Although the mtDNA sequence diversity (0.994 ± 0.001) and the mean pairwise differences (8.102 ± 3.774) in Cubans are relatively high, the WIMP value is low (2.869), suggesting that the sample set is composed of distantly related haplogroups with low to moderate internal diversity. According to the geographical origin attributed to each mtDNA haplogroup, 45% of the mtDNA sequences found in Cubans are of African origin, 33% of Native American origin and 22% of the lineages are of West Eurasian origin (namely, Europe and the Middle East) (Table 1).Within the African lineages, the vast majority of sequences belong to sub-Saharan L haplogroups (43.3% of the total sample), whereas a small proportion (2% of the total sample) fall into the typical North African U6 haplogroup. Interestingly, most of these U6 sequences (see Additional file 2) belong to the sub-clade U6b1, which is characteristic of Canary Islands [24]. The main U6b1 profile A16163G T16172C A16219G T16311C (three matches in Cuba) is in fact highly prevalent in the Canary Islands (~10%) [24], and outside these islands, it appears only sporadically in some other Latin-American countries such as Uruguay

Quote:

 

U6b1: found only in the Iberian peninsula and the Canary Islands. Estimated age: c. 6000 BP.

Quote:

 

U6a: it is the most widespread (from Canary Islands and Iberian Peninsula to Syria, Ethiopia and Kenya) and has highest diversity in Eastern Africa. Estimated age: 24-27,500 BP

Quote:

 

With respect to the Y-chromosome haplogroups, 78.8% of the chromosomes analyzed can be traced back to the West Eurasian gene pool, whereas the African fraction accounts for 19.7% of Cuban lineages (Table 2); two individuals (1.5%) carried Y-chromosomes of East Asian origin. Among the West Eurasian fraction, the vast majority of individuals belong to West European haplogroup R1 (xR1a). The African lineages found in Cubans have a Western (haplogroups E1, E2, E3a) and Northern (haplogroup E3b2) African origin. Interestingly, we did not observe Native American Y-chromosome lineages, such as those belonging to haplogroups P or Q [28]. Considering that our sample size of Y-chromosomes is 132 individuals (n), the highest frequency (F) of whatever unobserved haplogroup in the population, with a 95% probability, could be estimated as 1-e-Fn = 0.95, according to the Poisson distribution. Therefore, it is still possible that Native American haplogroups could be present in the Cuban population with roughly a maximum frequency of 2.3%.

Quote:

 

E1b1b1a (E-M78), formerly E3b1a, is a commonly occurring subclade, widely distributed in North Africa, the Horn of Africa, West Asia, i.e. The Middle East and Near East "up to Southern Asia",[Note 7] and all of Europe.[Note 8] The European distribution has a frequency peak centered in parts of the Balkans (up to almost 50%[3][21]) and Italy and declining frequencies evident toward western, central, and northeastern Europe.

Based on genetic STR variance data, Cruciani et al. (2007) suggests that this subclade originated in "Northeastern Africa", which in the study refers specifically to Egypt and Libya.[Note 9][Note 10] about 18,600 years ago (17,300 - 20,000 years ago).[Note 11] Battaglia et al. (2008) describe Egypt as "a hub for the distribution of the various geographically localized M78-related sub-clades" and, based on archaeological data, they propose that the point of origin of E-M78 (as opposed to later dispersals from Egypt) may have been in a refugium which "existed on the border of present-day Sudan and Egypt, near Lake Nubia, until the onset of a humid phase around 8500 BC. The northward-moving rainfall belts during this period could have also spurred a rapid migration of Mesolithic foragers northwards in Africa, the Levant and ultimately onwards to Asia Minor and Europe, where they each eventually differentiated into their regionally distinctive branches". Towards the south, Hassan et al. (2008) also explain evidence that some subclades of E-M78, specifically E-V12 and E-22, "might have been brought to Sudan from North Africa after the progressive desertification of the Sahara around 6,000-8,000 years ago".

Quote:

 

E1b1b1b (E-M81), formerly E3b1b or E3b2, is the most common Y chromosome haplogroup in the Maghreb, dominated by its sub-clade E-M183. It is thought to have originated in the area of North Africa 5,600 years ago.[2][22] It is colloquially referred to as the "Berber marker" for its prevalence among Mozabite, Moyen Atlas, Kabyle and other Amazigh groups, E-M81 is also quite common among North African Arab groups. It reaches frequencies of up to 80% among some groups of Morocco. This includes the Saharawish for whose men Bosch et al. (2001) reports that approximately 76% are M81+.
In Europe, E-M81 is found everywhere but mostly in the Iberian Peninsula, where unlike in the rest of Europe[Note 12] it is more common than E-M78, with an average frequency of 4-5.6%, and frequencies reaching 9% in Galicia, 10% in Western Andalusia and Northwest Castile and 9% to 17% in Cantabria.[24][25][26][27][28] The highest frequencies of this clade found so far in Europe were observed in the Pasiegos from Cantabria, ranging from 18% (8/45)[28] to 41% (23/56)[2].

Quote:

 

E1b1b1 (E-M35), formerly E3b1c or E3b3, is mostly known for its major sub-clade E1b1b1c1 (E-M34), which dominates this clade.[Note 14] However, earlier studies did not test for E-M34.

Attached to this post:

mt_DNA_1.jpg (52.38 KB)

mt_DNA_2.jpg (27.74 KB)

Y_DNA.jpg (35.1 KB)

Attached to this post:

Cuban_DNA_Project_Y_DNA_Haplogroups.jpg (28.19 KB)

Cuban_DNA_Project_Mt_DNA_Haplogroups.jpg (36.13 KB)

Quote:

 

A Study of the Genetical Structure of the Cuban Population:
Red Cell and Serum Biochemical Markers
R. GONZALEZ,1 J. M. BALLESTER,1 M. ESTRADA,1 F. LIMA,1 G. MARTINEZ,1
M. WADE,1 B. COLOMBO,1 AND R. VENTO2
INTRODUCTION
The present investigation is a genetic analysis of the Cuban population living in the
Havana province. It is well known that this population originated from Caucasian
Spaniards and African Negroes [1]. The original Cuban Indian population, which
consisted of approximately 100,000 individuals living in a neolithic society [2, 3],
was completely exterminated soon after the Spanish invasion beginning in 1510.
This rapid disappearance, which took place in less than 30 years, was the result of
epidemics, infant mortality, hunger, and many individual and collective suicides.
The suicides were the consequence of the psychological impact of Western civilization
for which "the submission to the insane, monotonous, and hard gold mine work,
without tribal meaning and without religious ritualism took away the reason for
life" [4]. Moreover, thousands of Indians were massacred at the beginning of the
conquest in attempts by the Spanish to subdue them and later to break their resistance;
in 1550 less than 5,000 had survived [5, 6]. Today approximately 500
Indians live in the most isolated mountains of the Oriente province. It was to this
province that their ancestors escaped and survived in complete isolation (Gonzalez
et al., in preparation). The contribution, therefore, of the Indian population to the
Cuban gene pool is negligible, so that the actual gene pool consists only of African
and European genes.
Previously, racial classification of Negroes, mulattoes, and whites, based only on
morphological appearance (i.e., color of skin, appearance of hair, and thickness of
lips, etc.), was used in population genetics and in the study of gene flow between
Caucasians and Negroes [7, 8]. Racial interbreeding in the Cuban population, for
example, is revealed by the variation of the morphological characteristics. More
recently, the development of biochemical techniques for the determination of welldefined
genetic characters-many of which allow the analysis of the primary gene
product-has permitted the study of the true genetic makeup of a population
We report a study of the genetic polymorphisms in Negroes, Caucasians, and mulattoes in the city of Havana. The results were used to measure the genetic
admixture between the two original groups (i.e., the amount of Caucasian and
Negro genes in the mulatto group) and to calculate gene frequencies in the general
Cuban population.
THE SAMPLE
The sample was arbitrarily made up of whites, mulattoes, and Negroes randomly
chosen from all the donors in the Blood Bank of Havana.
The classification of the three groups was based on the following morphological aspects:
colour of the skin, appearance of the hair, thickness of the lips, nose type, and hairiness.
These features could not be quantitated, but the same trained person classified all the
individuals on the basis of his personal evaluation. To minimize error, each individual was
asked about his racial ancestry.
The morphological variability in the mulatto group was very wide, but instead of
dividing this group into subgroups as done by other investigators [7], we included
"dark" as well as "light" mulattoes.
The sample consisted of 155 whites, 145 mulattoes, and 168 Negroes; all were males.
METHODS
Blood was collected in heparinized test tubes for enzyme determinations and without
anticoagulant for serum analysis. Blood cells were immediately separated by centrifugation
and washed three times with physiological saline solution at 40C; packed cells were stored
at -20'C and examined within a week. Serum was separated from the coagulated blood
cells by centrifugation and stored at -200C.
The following genetic markers were studied using the techniques described in the corresponding
references: glucose-6-phosphate dehydrogenase (G6PD) [12]; 6-phosphogluconate-
dehydrogenase (6PGD) [13]; adenosine deaminase (ADA) [14]; phosphohexose
isomerase (PHI) [15]; acid phosphatase (ACP) [16]; phosphoglucomutase (PGM1)
[17]; adenylate kinase (AK1) [18]; peptidase A (PEPA) [19]; hemoglobin phenotypes
(Hb) [20]; haptoglobin (Hp) [21]; diaphorase (DIA1) [22]; lactate dehydrogenase
(LDH) [22].
G6PD activity was determined in all cases showing a GdA phenotype to discriminate
between GdA and GdA- alleles [12].
RESULTS
PHI and LDH did not show any detectable polymorphism; in both cases one
rare (slow) variant was found. Table 1 shows the phenotypes and the gene frequencies
found for the other 10 markers studied in this survey. It is worth noting
that all cases for which the Hardy-Weinberg equilibrium could be tested showed
the expected frequencies.
DISCUSSION
The frequencies for 12 genetic traits in samples from the Havana Blood Bank
were used to characterize the genetic makeup of the Caucasians and Negroes living
in Cuba. Because these two groups originated in Spain and West Africa respectively,
frequencies found in the Caucasians are compared with frequencies of the
same traits in contemporary Spaniards, and those of the Negroes, with contemporary
West African Negroes in table 2. As expected, an increasing African contribution from Caucasians to WVest African Negroes, going through the three
Cuban groups, was indeed observed.
These data can be used to measure the gene flow between these two groups with
the following restrictions: (1) the estimates of the gene frequencies of the base
populations are only approximate because it is impossible to locate exactly the
places from which present day groups originated; (2) present frequencies are the
same as those existing several generations ago; (3) selection might have occurred
together with admixture; and (4) the problems of sampling (see Reed [311).
Estimates of gene migration were calculated using the method of Bernstein [32];
the errors were calculated according to Reed [31]. The results for the alleles
considered are given in table 3.
Data obtained in the African regions from which Cuban Negroes probably
originated were used for the calculations. The heterogeneity of the ancestral
population brought to Cuba is well known [ 1 ], and the importance of such
heterogeneity on the admixture estimates has been discussed [331. Unfortunately,
the extent of this heterogeneity cannot be calculated because no reliable data are
available. Historical records are scanty and biased for the following reasons
[34-36]: (1) many Negro cargoes were smuggled so that a considerable number
of arrivals were not recorded; (2) the identification of the slaves was based on
the port of embarkment, not on the place of origin; (3) generally the composition
of the slaves carried in one ship was very heterogeneous to avoid mutinies; and
(4) the ethnographic identification was based mainly on the efficiency, character,
and submissiveness of the slave; thus, the slave traders used to sell one type of
slave for another with preference given to the most efficient. Considering these
observations and the average gene frequencies for the United States Negro given
by Adams and Ward [33], which are very similar to those used to calculate the genetic admixtures, we concluded that it is not possible to consider this heterogeneity,
and moreover it would not substantially improve the estimates of M.
In spite of this and the previously mentioned sources of uncertainty, the genes
analyzed seem to provide a rather consistent estimate of the genetic admixtures
studied. Moreover, these data indicate good agreement between the genetic composition
of each group and the discrimination based on morphological characteristics.
As expected [30], the estimates of Negro contribution to the three groups
studied, based on /', turned out to be lower than those obtained with other
markers. The most likely explanation for this finding is that selection acted against
this gene. The records available on the presence of malaria in Cuba indicate that
the endemic form was localized in the eastern part of the country (Oriente and
Camaguey provinces) and that Havana province has been nonmalarious-at least
since 1920 [37]. Although records during the eighteenth and nineteenth centuries
are poor and incomplete, they too suggest that the Havana province was never
affected by the endemic form of this disease [38].
The only marker which gave apparently inconsistent results was ACP, as can be
seen in table 3; the reasons for this finding are unknown, but a similar discrepancy
has been already reported by Terrenato et al. [39] in a study carried out to
determine the genetic distances among several Sardinian villages using many
genetic markers.
We can then conclude that a reasonably reliable estimate of the Caucasian and
African genetic admixture in the three groups studied (using only GdA, GdA,
PEPA2 and Hp') is the following: (1)Cuban Caucasians: 5% Negro genes (see
column 1, table 3) and therefore 95%, Caucasian genes; (2) Cuban mulattoes:
53 % Caucasian genes (see column 2, table 3) and therefore 47 % Negro genes;
592 (3) Cuban Negroes: 13% Caucasian genes (see column 4, table 3) and therefore
87% Negro genes.The estimates of these mean values (M) were calculated according to Cavalli-
Sforza and Bodmer [40]. In figure P-the comparison between M and the estimates
of the African contribution to the genetic makeup of the three groups studied,
obtained for each allele, is reported.
586

Attached to this post:

Fig_1.jpg (35.49 KB)

Table_2.jpg (60.17 KB)

Table_3.jpg (52.73 KB)

Delilah

Jul 28 2010, 04:08 AM

Abnormal Hemoglobins and Racial Admixture in Villa Clara Province (1999)
Quote:
Summary:

Frequencies of the HBB*S and HBB*C alleles were estimated on the basis of a sample of 6007 Whites and 5084 Negroids from Villa Clara, Cuba. The allele frequencies estimates in Negroids were HBB*S = 0.0306 and HBB*C = 0.0030. As much as 54% of the alleles present in the Negroids may be of Europid origin, while persons classified as White may have from 2 to 7% of African ancestry.

You might know of this one Basque. It's old. I read a similar thing about how White Brazilians often also have Negroid ancestry. Of course not as much as those classified as mixed. But its there regardless. Sometimes I wonder if its due to real admixture or its has to do with the statistical ways they use to calculate admixture. Did you see what I posted about the Puerto Ricans and how a subsequent study was not able to replicate the earlier findings of increased African ancestry at Chromosome 6? Something called linkage disequilibrium affected the results.

Oh and actually these results are not so different from us mainly European Puerto Ricans. The only thing is that Puerto Ricans who have between 88% and 92% European also have Asian admixture. This lowers further their European percentages. Cubans have very little Native-American admixture.


There seems to have been a problem with the above study. It was discussed here: http://thestudyofracialism.org/viewtopic.php?t=1771

Quote:

 

fwsweet
Administrator
Here are some more studies on Afro-European genetic admixture in Cubans that you may find useful:
Hanis 1991
Gonzalez 1975
Carpio 2005
Torroni 1995
Gonzalez 1976

Of the above studies, the first one (Hanis 1991) is probably the one most worth getting a copy of. It may seem irrelevant because it is about susceptibility to diabetes, but it summarizes and tabulates admixture measurements from many prior studies. Consequently, it shows the measurement ranges of many different studies. The summary of all the studies is as follows:

Mexican-Americans are 61 percent European, 31 percent Native American and 8 percent African. Puerto Ricans in the U.S. are 45 percent European, 18 percent Native American, and 37 percent African. Cuban-Americans are 62 percent European, 18 percent Native American, and 20 percent African.

The Gonzalez 1975 study was more limited and did not try to distinguish Native American from European. For Havana, it found 80 percent Euro-Indian and 20 percent African.

Quote:

 

Abstract
The purpose of this article was to characterize the origins of the United States Hispanic population and discuss the implications of these origins in the context of diabetes risk. Particular attention was focused on the genetic origins of the three major U.S. Hispanic groups, i.e., Mexican Americans, Puerto Ricans, and Cubans. The U.S. Census figures provided basic demographic information. Genetic marker data for ancestral populations were taken from a review of the literature and compendia. Genetic marker data for the Puerto Rican and Cuban populations were extracted from the literature. Genetic markers determined on approximately 1000 randomly selected Mexican Americans from Starr County, Texas, were taken as representative of the Mexican-American population. The Hispanic population is the second largest and fastest growing minority in the U.S. Estimates of the Hispanic population in 1988 indicated some 19.4 million residents, of whom 62% were classified as Mexican, 13% as Puerto Rican, and the remaining 25% as Cubans and others. Various lines of evidence can be used to characterize the Hispanic population and its origins. These include ethnohistory, self-assessment of ancestry, surname distributions, speech and cultural characteristics, quantitative traits, and genetic structure. Genetic data were used to estimate the contribution of putative ancestral populations to the contemporary gene pool. For Mexican Americans, 31% of the contemporary gene pool is estimated to be Native American derived, whereas 61 and 8% are Spanish and African derived, respectively. In Puerto Rico, the percentage of contributions of Spanish, Native American, and African admixture to the population are 45, 18, and 37%, respectively. For Cuba, the parallel estimates are 62, 18, and 20%. The high frequency of Native American-derived genes in the contemporary Hispanic population predict a higher frequency of non-insulin-dependent diabetes mellitus (NIDDM) under the assumption that genes are important in NIDDM etiology. Our results are consistent with the finding of the significant role of genes in determining risk.

Quote:

 

The Gonzalez 1975 study was more limited and did not try to distinguish Native American from European. For Havana, it found 80 percent Euro-Indian and 20 percent African.

Quote:

 

Today approximately 500
Indians live in the most isolated mountains of the Oriente province. It was to this
province that their ancestors escaped and survived in complete isolation (Gonzalez
et al., in preparation). The contribution, therefore, of the Indian population to the
Cuban gene pool is negligible, so that the actual gene pool consists only of African
and European genes.
[...]
THE SAMPLE
The sample was arbitrarily made up of whites, mulattoes, and Negroes randomly
chosen from all the donors in the Blood Bank of Havana.
The classification of the three groups was based on the following morphological aspects:
colour of the skin, appearance of the hair, thickness of the lips, nose type, and hairiness.
These features could not be quantitated, but the same trained person classified all the
individuals on the basis of his personal evaluation. To minimize error, each individual was
asked about his racial ancestry.
The morphological variability in the mulatto group was very wide, but instead of
dividing this group into subgroups as done by other investigators [7], we included
"dark" as well as "light" mulattoes.
The sample consisted of 155 whites, 145 mulattoes, and 168 Negroes; all were males.
We can then conclude that a reasonably reliable estimate of the Caucasian and
African genetic admixture in the three groups studied (using only GdA, GdA,
PEPA2 and Hp') is the following: (1) Cuban Caucasians: 5% Negro genes (see
column 1, table 3) and therefore 95%, Caucasian genes; (2) Cuban mulattoes:
537% Caucasian genes (see column 2, table 3) and therefore 47% Negro genes;
59 % (3) Cuban Negroes: 13%o Caucasian genes (see column 4, table 3) and therefore
87%o Negro genes.

Quote:

 

The first misunderstanding – You wrote that the 2-7 percent of African admixture measured in “white” Cubans in the Villa Clara study is not what you would call “high.” In my view, it is very high indeed—too high. It is ten times the African admixture found in White USAmericans (which is about 0.7 percent). Recall that this study measured the African admixture only in people who were “classified as white.” That is, it looked only at people having no discernable African features. Despite people like Mark Shriver (utterly European-looking despite significant African admixture) and Rick Kittles (utterly African-looking despite significant European admixture) who demonstrate the luck of the draw difference between external features and internal admixture, over the long pull it is safe to say that in most people, most of the time, the more African admixture you have the more African you look. The break point for USAmericans is around 15-20 percent. People with more admixture than that look at least partially “negroid” to most Americans. And so, if you deliberately choose a population comprising only people who “look white,” the maximum within that population can be no more than 17 percent, the median will be around 8 percent, and the mode and mean will be less than that, probably around 4 percent. That they found a maximum of 7 percent seems suspiciously high to me. That they found a minimum of 2 percent is simply incredible; the minimum should have been zero.