Genetic variation of 15 autosomal STR loci in Upper (Southern) Egyptians

Population: A sample of 265 unrelated volunteers from five governorates (El-Minia, Assiut, Sohag, Qena and Aswan) in Upper (south) Egypt (Fig. 1), was collected. After obtaining appropriate ethical approval, the participants were questioned for their ethnic origin as inhabitants of Upper Egypt for at least the third generation and they signed informed consents. The residents are culturally and linguistically distinguishable from those of the Northern population, where people speak an Arabic language with a distinctive dialect.

View Large Image Download to PowerPoint

Fig. 1. Map of Egypt (26 governorates), showing the geographical location of Upper Egypt's governorates (14–21) with those included in the present study underlined; adapted from: http://lcweb2.loc.gov/frd/cs/egypt/eg00_06a.pdf.

Extraction: Genomic DNA was extracted from blood stains and buccal swabs by a modified silica-based method [1]. DNA quantification was undertaken using the Quant-iT™ High Sensitivity DNA Assay Kit (Molecular Probes) according to the manufacturer's instructions.

PCR: 0.5–1ng DNA templates were amplified in half quantities according to the manufacturer's instructions (AmpFℓSTR® Identifiler™ PCR Amplification Kit: User's Manual, Applied Biosystems).

Typing: The PCR products were detected with the ABI PRISM® 3100 Genetic Analyzer (Applied Biosystems) and sized with GeneScan-500 LIZ internal lane size standard. Allelic calls and genotyping were carried out by comparison to the reference allelic ladder included in the kit, using GeneMapper ID® v3.2 (Applied Biosystems).

Quality control: Identifiler™ kit included allelic ladder and positive control.

Analysis of data: Microsoft Excel-PowerStats v1.2 [2] was used for allele frequency and forensic parameters calculations including heterozygosity (H), polymorphic information content (PIC), power of discrimination (PD), power of exclusion (PE) and matching probability (MP) for each locus of the studied population. Hardy–Weinberg equilibrium (HWE), Linkage disequilibrium (LD), population differentiation tests, FST genetic distances and pair-wise analysis of molecular variance (AMOVA) for comparison with other populations, were carried out using Arlequin v3.01 [3] and GenAlEx 6 software [4]. In addition, graphical representation of genetic distances (FST distances) using a two-dimensional genetic map based on multi-dimensional scaling (MDS) analysis was performed by means of SPSS v14.0 software. Both frequency data gathered from literature and actual genotypic data obtained through personal communications, were considered for comparative analyses with other populations at the local, regional and global levels (Table 1) [5], [6].

Table 1. Diverse (global) populations and constituent subpopulations included in the genetic distance analyses with Upper Egyptians

	Population	Number	Code	Contained subpopulations
	South Asian	198	SAS	Pakistani (Brahui, Balochi, Hazara, Makrani, Sindhi, Pathan, Kalash, Burusho)
	East Asian	250	EAS	Japan, China (Han, Tuji, Yizu, Miaozu, Orogen, Daur, Mongolia, Hezhen, Xibo, Uygur, Dai, Lahu, She, Naxitu), Siberia (Yakut)
	Sub-Saharan African	125	SSA	Central African Republic (Biaka), Congo (Mbuti), Senegal (Mandenka), Nigeria (Yoruba), Namibia (San), South Africa (Bantu), Kenya (Bantu)
	European	158	EUR	France (Basque), Italy (Sardinian, Bergamo, Tuscan), Russia (Caucasus, Adygei), Orkney Islands
	Oceanian	39	OCE	New Guinea (Papuan), Bougainville (Melanesian)
	American	108	AM	Colombia, Brazil (Surui, Karitiana), Mexico (Maya, Pima)
	North African	30	NAF	Algeria (Mozabites)
	Middle Eastern	148	ME	Israel (Druze, Bedouin, Palestinian)

Access to the data: Supplementary data available online as an Excel file.

Results: Allele frequency data of the 15 STR loci are presented in Table 2. HWE evaluation and forensic parameters are included in Table 3. Assessment of genetic distances (FST) based on frequency data comparison with other local and regional population versus Upper Egyptians is collected in Table 4. Multi-dimensional scaling plot derived from the FST genetic distances among Upper Egyptians and other populations compared globally is shown in Fig. 2.

Table 2. Allele frequencies of 15 STR loci included in the Identifiler kit in Upper Egyptians

	Allele	D8	D21	D7	CSFIPO	D3	TH01	D13	D16	D2	D19	vWA	TPOX	D18	D5	FGA
	5			0.002
	6						0.206						0.009
	7			0.011	0.008		0.204						0.006
	8	0.015		0.16	0.021		0.115	0.126	0.036				0.483	0.002	0.051
	9	0.006		0.094	0.015		0.374	0.094	0.151				0.211		0.081
	9.3						0.081
	10	0.057		0.342	0.264		0.019	0.047	0.104		0.008		0.062	0.002	0.125
	10.2													0.002
	11	0.108		0.245	0.311		0.002	0.226	0.268		0.011		0.217	0.008	0.225
	11.2										0.002			0.009
	12	0.111		0.123	0.338			0.338	0.255		0.113		0.011	0.166	0.364
	12.2										0.008			0.004
	13	0.208		0.021	0.036	0.006		0.119	0.162		0.228	0.002		0.104	0.145
	13.2										0.034			0.002
	14	0.242		0.002	0.008	0.072		0.047	0.025		0.247	0.087		0.128	0.009
	14.2										0.053			0.002
	15	0.191				0.306		0.002			0.143	0.104		0.128
	15.2										0.057			0.002
	16	0.046				0.262				0.04	0.053	0.292		0.134
	16.1															0.002
	16.2										0.04			0.002
	17	0.011				0.228				0.232	0.004	0.272		0.102		0.002
	17.2													0.002
	18	0.004				0.117				0.136		0.164		0.074		0.011
	18.2													0.002
	19					0.009				0.113		0.066		0.062		0.049
	20									0.151		0.013		0.03		0.077
	20.2															0.004
	21									0.062				0.026		0.134
	21.2															0.004
	22									0.062				0.008		0.192
	22.2															0.006
	22.3															0.002
	23									0.092						0.158
	23.2															0.002
	24									0.058						0.157
	25									0.043						0.098
	26		0.004							0.008						0.057
	27		0.036							0.002						0.023
	28		0.123													0.015
	28.2		0.004
	29		0.268													0.006
	29.2		0.002
	30		0.243													0.002
	30.2		0.013
	31		0.047
	31.2		0.089
	32		0.004
	32.2		0.106
	33		0.002
	33.2		0.04
	34		0.004
	34.2		0.006
	35		0.006
	36		0.002
	37		0.002
	38		0.002

The highest allele frequency for each locus is in bold.

Table 3. Forensic parameters and the exact test probability values for Hardy–Weinberg equilibrium evaluation

	Locus	D8	D21	D7	CSFIPO	D3	TH01	D13	D16	D2	D19	vWA	TPOX	D18	D5	FGA
	MP	0.05	0.051	0.087	0.133	0.1	0.096	0.075	0.069	0.033	0.045	0.076	0.16	0.026	0.09	0.033
	PEX	0.627	0.693	0.558	0.473	0.565	0.431	0.524	0.592	0.753	0.565	0.48	0.37	0.664	0.551	0.807
	PD	0.95	0.949	0.913	0.867	0.9	0.904	0.925	0.931	0.967	0.955	0.924	0.84	0.974	0.91	0.967
	PIC	0.81	0.81	0.74	0.66	0.73	0.73	0.76	0.77	0.86	0.82	0.76	0.62	0.88	0.74	0.86
	H (%)	81.5	84.9	77.7	72.5	78.1	70.2	75.8	79.6	87.9	78.1	73.2	66	83.4	77.4	90.6
	Ho	0.815	0.849	0.777	0.728	0.781	0.706	0.758	0.796	0.879	0.781	0.732	0.660	0.834	0.774	0.906
	He	0.834	0.831	0.777	0.720	0.768	0.757	0.794	0.805	0.871	0.844	0.793	0.673	0.891	0.795	0.874
	p	0.165	0.947	0.489	0.219	0.105	0.182	0.337	0.459	0.861	0.013	0.119	0.414	0.000	0.000	0.341

MP, match probability; PEX, power of exclusion; PD, power of discrimination; PIC, polymorphism information content; H, heterozygosity; Ho, heterozygosity observed; He, heterozygosity expected. Significant p values are in bold.

Table 4. Genetic distances (FST) p-values between Upper Egyptians and other related populations.

	Loci	U.Eg vs. Eg.El-Minia	U.Eg vs. Eg.Christians	U.Eg vs. Eg. Cairo	U.Eg vs. Dubai	U.Eg vs. Is. Jewish	U.Eg vs. Is. Arabs	U.Eg vs. Syrian	U.Eg vs. Moroccan	U.Eg vs. Qatari	U.Eg vs. Omani	U.Eg vs. Tunisean	U.Eg vs. Algerian	U.Eg vs. Yemenite	U.Eg vs. Turkish	U.Eg vs. Greek	U.Eg vs. Italian (NW)
	D8S117	–	0.578	0.991	0.431	–	–	0.207	0.668	0.334	0.476	0.199	0.142	0.145	0.019	0.847	0.000
	D21S11	–	0.849	0.358	0.074	–	–	0.570	0.811	0.636	0.209	0.991	0.617	–	0.537	0.674	0.701
	D7S820	0.907	–	0.180	0.950	0.189	0.153	0.936	0.199	0.579	0.999	0.828	0.567	0.939	0.038	0.513	0.567
	CSFIPO	0.456	–	–	0.263	0.377	0.352	0.622	0.994	0.988	–	0.000	0.605	–	0.853	–	0.721
	D3S1358	0.672	0.729	0.769	0.516	–	–	0.651	0.826	0.147	0.000	0.893	0.808	0.179	0.475	0.080	0.422
	TH01	0.001	–	–	0.000	0.000	0.000	0.028	0.336	0.045	–	0.001	0.439	–	0.000	–	0.000
	D13S317	O.2C0	0.991	0.447	0.255	0.011	0.736	0.056	0.217	0.102	0.440	0.008	0.499	0.283	0.007	0.145	0.493
	D16S539	0.633	–	–	0.571	0.219	0.183	0.914	0.433	0.004	–	0.298	–	–	0.584	–	0.C97
	D2S1338	–	–	–	–	–	–	–	–	0.857	–	–	–	–	0.719	–	–
	D19S433	–	–	–	–	–	–	–	–	0.023	–	–	–	–	0.216	–	–
	vWA	0.368	0.685	0.968	0.434	0.121	0.895	0.726	0.508	0.567	0.668	0.419	0.142	0.647	0.056	0.045	0.025
	TPOX	0.960	–	–	0.086	0.061	0.155	0.000	0.000	0.000	–	0.000	0.000	–	0.000	–	0.000
	D18S51	–	0.977	0.192	0.336	–	–	0.441	0.527	0.071	0.341	0.267	0.638	–	0.022	0.373	0.871
	D5S818	0.535	0.592	0.288	0.153	–	–	0.029	0.164	0.134	0.372	0.128	0.169	0.793	0.032	0.307	0.129
	FGA	0.934	0.449	0.223	0.808	–	–	0.249	0.832	0.000	0.823	0.763	0.052	0.643	0.252	0.984	0.091

Numbers in bold indicate significant (FST) p-values (significance level is 0.05). U.Eg, Upper Egyptians; IS, Israeli; NW, North West.

View Large Image Download to PowerPoint

Fig. 2. Multi-dimensional scaling (MDS) based on pair-wise FST genetic distances of Upper Egyptian and other diverse global populations. OCE, Oceanian; ME, Middle Eastern; NAF, North African; EAS, East Asian; SSA, sub-Saharan African; UEGY, Upper Egyptian; SAS, South Asian; EUR, European. The figure shows that Oceania and American populations are very distant from Upper Egyptians (marked by a grey triangle) and other populations. The Upper Egyptian population is closer to the Middle Eastern, North African, South Asian and European populations than others.

Other remarks: Pair-wise FST genetic distances between sampled subpopulations from Upper Egypt revealed no significant differences (data not shown), so they have been treated as a single population in subsequent analyses. The frequencies of alleles at all loci along with forensic parameters were addressed (Table 2, Table 3). All loci are highly polymorphic with the highest being D18S51 (0.88), and the lowest is TPOX (0.62). The most polymorphic loci were found to be the most discriminating as well. The combined power of exclusion was calculated to be 99.999868% and the combined match probability was 1.93×10−18. Hardy–Weinberg Equilibrium expectations were met for all loci except for D18S51, D5S818 and D19S51 (p-values: 0.000, 0.000 and 0.013, respectively) (Table 3). However, the exact test p-values for the first two loci remained significant even after Bonferroni's correction. Possible explanations for HWE departure are population substructure and admixture, since there is persistent consanguinity (27.7% in WHO report, 1994) [7] in addition to historical admixture with other ethnic groups who ruled Egypt in its past, e.g. Ottomans, Arabs (who constituted major migration waves) and Greeks. There was no evidence of departure from independence of alleles between all tested loci (linkage equilibrium).

Local comparisons between Upper Egyptians were carried out with other ethnic groups in Egypt, based on frequency and molecular data. No differences were observed in comparison with a general Caucasian population from Cairo in any of the nine loci compared [8] or with Egyptian Christians from Cairo [9], but one out of eight loci showed a difference in comparison with a population from El-Minia city [10]. At the molecular data level, there was a weak significant difference when Upper Egyptians were compared with Egyptian Muslims from Tanta (FST p-value=0.0455), albeit with a non-significant p-value (0.143) in an exact test of population differentiation. Highly significant differences (p<0.00001) were observed in comparisons with Berbers from Siwa (an oasis in the western desert), and with a population sample from Adaima (a village in Upper Egypt) [11].

Global and regional comparisons, based on genotypes of all loci except D2S1338 and D19S433, with diverse populations from South Asia, sub-Saharan Africa, Europe, Oceania, Middle East, the Americas and North Africa were performed to address relationships with neighbouring and distant populations. A highly significant FST genetic difference was observed with all groups (p-values <0.00001). The genetic relationships with these populations were demonstrated in a multi-dimensional scaling plot which showed that Upper Egyptians are located centrally in relation to other continental groups. They were very distant from Oceanian and American populations and clustered with Middle Eastern populations (including Israeli Druze, Bedouins and Palestinians) rather than with Europeans, South Asians and North Africans (Fig. 2).

Further comparisons of genetic distances were undertaken with other related Arab and European populations by means of published allele frequency data [12], [13], [14], [15] (Table 4). These showed the following significant differences: 5 out of 15 loci in Qatari; three and one out of 13 loci in Syrians and Moroccans, respectively; 1 and 4 out of 12 loci in Algerian Mozabites and Tunisians, respectively; 1 out of 9 loci in Omani; 2 out of 7 in an Israeli Jewish sample. No difference was observed with Yemenites in the seven loci examined. Regarding the European populations, Turkish, Greeks and Italians, significant dissimilarity was found in 7 out of 15 loci, 1 out of 9 and 4 out of 13, respectively [16], [17], [18].

Clustering of Upper Egyptians with Middle Eastern populations, and weak differentiation from most other Arab populations, is expected because of geographical proximity and gene flow to Egypt, which has acted as a crossroad to the three continents throughout its history. This observation is supported by mitochondrial DNA and Y chromosome studies which have suggested a regional genetic continuity among populations of the Nile valley, Middle East and Arabian Peninsula [19], [20]. This paper follows the guidelines suggested for publication of population data [21].