Rapidly Mutating Y Chromosomal STRs in Deep-Rooted Endogamous Pedigrees

Y-chromosome short tandem repeat polymorphisms (Y-STRs) are important in many areas of human genetics. Y chromosomal STRs being normally utilized in the eld of forensic exhibit low haplotype diversity in endogamous populations and fail to discriminate among male relatives from same pedigree. Rapidly mutating Y-STRs (RM Y-STRs) have been paid much attention in last decade. These 13 RM Y STRs have high mutation rates (>10-2) and have considerably higher haplotype diversity and discrimination capacity than conventionally used Y-STRs showing remarkable power, when it comes to differentiation in paternal lineages in endogamous populations. Previously, we have analyzed 2–4 generation, 99 pedigrees covering 1568 pairs of men covering 1–6 meioses from all over Pakistan and 216 male relatives from 18 deep rooted endogamous Sindhi pedigrees covering 1-7 meioses. Here we are presenting 861 pairs of men from 63 endogamous pedigrees covering 1-6 meioses from Punjabi population of Punjab, Pakistan. Mutations were frequently observed at DYF399 and DYF403 while no mutation was observed at DYS526a/b. The rate of differentiation ranged from 29.70% (rst meiosis) to 80.95% (fth meiosis) while overall (1 to 6 meiosis) differentiation was 59.46%. Combining previously published data with newly generated data, an overall differentiation rate was 38.79% based on 5176 pairs of men related by 1–20 meioses, while Y-ler differentiation was 9.24% based on 3864 pairs. Using father-son pair data from the present and previous studies, we also provide updated RM Y-STR mutation rates. based on a total number of 878 47731 meioses.

Introduction Y-chromosomal microsatellites or short tandem repeats (Y-STRs) play an important role within forensic genetics [1][2][3] . Most commonly, Y-STRs are used to determine the male component of DNA mixtures when a high female background is present 4,5 , or to determine paternal relationships between male individuals 6-8 . Currently, a large and growing reference databases exists for estimating Y-STR haplotype frequencies among worldwide human populations (e.g. http://www.yhrd.org or http://usystrdatabase.org/). Commercially available Y-STR kits (Y ler, Powerplex 23 and Y ler Plus) are valuable, but they have some limitations for their use in forensic casework. Overall, the haplotype diversities (HD) of these sets of Y STRs are good for outbreeding populations, usually at 0.995 and higher 9,10 . However, the ability to discriminate between individuals is much lower than that of the autosomal STRs. In endogamous populations or populations which run into size contraction followed by quick expansion recently 11,12 , or have particular cultural practices such as patrilocality 13,14 , the currently used Y-STR panels provide limited resolution due to the overall reduced Y-chromosome diversity.
The major drawback of currently used Y-STRs is that they are unable to exclude close or distant patrilineal relatives of the suspect from having deposited the biological material instead of the suspect himself. These STRs maybe helpful in those cases where close or distant male relatives may be involved because of their relatively low mutation rates of only a few mutations per thousand generations per locus 6,8,15,16 .
In 2010, Ballentyne et al 16 reported mutation rates of 186 Y STRs in 2000 father-son pairs. This study identi ed 13 Y-STR markers with markedly higher mutation rate of about 10 2 and termed them rapidly mutating (RM) Y-STRs.
In the current study, we have provided empirical evidence of the ability of the 13 RM Y-STRs for improving paternal lineage resolution by analysing 861 pairs of men from 63 endogamous pedigrees covering 1-6 meiosis from Punjab, Pakistan.

Pedigree Samples and DNA Extraction
Initially 63 pedigrees 1-4 generations consisting of 861 pairs of men were located in different areas of Punjab Pakistan. 327 Blood samples were collected from 63 families. Pedigree of these family members was generated according to their oral records, which were later on con rmed with National Identity Card (NIC). In case of any confusion family registration certi cate (FRC) was applied using NADRA (National Database and Registration Authority) online paid services. All participants gave their informed consent either orally (in case they could not write) or in writing, after the study aims and procedures were carefully explained to them. This study was approved by the ethical review board of University of Health Sciences Lahore Pakistan and was in accordance with the principles of the Declaration of Helsinki made at 64th WMA General Assembly, Fortaleza, Brazil, in October 2013. All blood samples were stored at -20 ºC before DNA extraction. DNA was isolated from blood using ReliaPrep™ PCR ampli cation and genotyping PCR co-ampli cation of thirteen rapidly mutating Y-STR loci (DYF387S1, DYF399S1, DYF403S1ab, DYF404S1, DYS449, DYS518, DYS526ab, DYS547, DYS570, DYS576, DYS612, DYS626 and DYS627) were performed in a 5-dye uorescence-based multiplex reaction using the RM-Yplex assay 17 . From 1 to 2 ng of the target DNA was ampli ed according to the manufacturer's recommended protocol. Thermal cycling was conducted under the following conditions: 95°C for 10 min; 20 cycles of 94°C for 30 s, 55°C for 45 s, 72°C for 60 s; and a nal extension of 72°C for 45 min. All loci were ampli ed in a GeneAmp PCR System 9700 thermal cycler (Applied Biosystems, Foster City, CA). Ampli ed products were analyzed regarding GS600 LIZ size standard and Allelic Ladder using an ABI 3500 genetic analyzer (Applied Biosystems, Foster City, CA) with the POP-6 TM polymer (Life Technologies). Samples were analysed using GeneMapper®ID-X software version 1.2 at a threshold of 50 rfu.

Statistical analysis
Rate of differentiation among male relative pairs were calculated as number of differentiated pairs of relatives by one or more Y STRs divided by the total number of male relative pairs on that particular meiosis or degree of relationship (i.e., pair members separated by 1-20 meiosis). Total numbers of mutations observed dived by number of father-son tested were used to calculate the mutation rates and binomial standard deviation was used to calculate the mutation rates 95% con dence intervals (CI) which is available via http://statpages.org/con nt.html. Haplotype diversities were calculated as where n is the number of samples, and the frequency of the ith haplotype.

Non-paternity Issues
The non-paternity events are common in pedigrees and these events were also observed in Pakistani population dataset. To overcome this issue, we split the pedigrees accordingly. In some pedigrees only one individual was involved in non-paternity. After measuring the genetic distance that individual was removed from pedigree. Individuals with mutation event on ≥ 4 STRs were removed. This threshold was derived from our previous observations based on >2000 father-son pair carrying mutations at three Y-STRs; while none of these showed mutations at more than four Y-STR markers 8,18 .
Merging previously published data and newly generated data showed an overall rate of 26.55% from a total of 2990 pairs (Table 3). In current study, we provide an update on male relative differentiation for father-son pairs (n = 726) relative to the previous studies (total n = 1460) 18-20 .
Combing RM Y-STR data (   Hopefully as more data is gathered for RM Y-STRs, these differentiation rates will become more robust, especially for 2nd degree male relatives (beyond 4th, 5th and 6th meioses).
We have shown this with the differentiation of father-son pairs from the rst to the present study. In rst study Ballantyne et al. 16 (Table 3). Nonetheless, conventional Y-STRs tends to have lower mutation rates 8,18 and 9.24% of all related males from the combined studies were differentiated with Y-ler as opposed to 38.79% with RM Y-STRs.

Mutation rate estimates from father-son pairs
We also have calculated the mutation rates of these rapidly mutating Y-STRs which are based on 135 father-son pairs extracted from the 63 pedigrees, and combined this data with previous studies 6, 15,16,20,21 , where the same rapidly mutating Y-STRs panel was implemented for other father-son pairs ( Table 4). The mutation rates ranged from 0 (0-2.70 X 10 −2 ) for DYS526a orDYS526b to 1.778 X 10 −1 (1.174 X 10 −1 to 2.529 X DYS626a to 7.45 X 10 −2 (6.58 X 10 −2 to 8.39 X 10 −2 ) for DYF399S1. These currently estimated mutation rates are most reliable for the 13 RM Y-STR markers, given the underlying number of meiosis ranging from 2949 (DYS570) to 3327 (DYF387S1) between markers. The average mutation rate across all 13 RM Y-STR markers was 1.84 X 10 −2 (1.72 X 10 −2 -1.96 X 10 −2 ) based on a total number of 878 mutations from 47731 meioses.

Population genetic analysis in the endogamous Punjabi men
Out of 861 pairs, 75 unrelated Punjabi men, all individuals carried a unique RM Y-STR haplotype with haplotype diversity (HD) 1. In a previous multicenter global study 21 21 based RM Y-STRs also reported meaningfully lower haplotype diversities and lower unique haplotype proportions in endogamous ethnic groups than in urban and rural groups 21 . Pakistani population is generally considered highly endogamous, and in our previous study 99 pedigrees were sampled from urban (N = 48) and rural (N = 51) areas. We didn't observe any effect on RM Y-STR diversity, which may be due to small sample size. In current study 75 pedigrees were sampled from urban (N = 46) and rural (N = 29) areas and again we didn't observe any effect on RM Y-STR diversity.

RM marker differentiation per pairs
We have calculated the differentiation power of 13 RM Y STRs in 75 pedigrees. On rst meiosis 59 pairs were differentiated out of 135 and DYF399S1 differentiated 24 pairs (40.67%) while DYS526 a/b did not differentiate any pair. This trend was also followed in other pairs where DYF399S1 differentiated most of the pairs while DYS526 a/b did not differentiate any of the pairs (Table 5). This pattern was concordant with previous studies 6, 15,20,21 . differentiation between paternally related males and across 4th meiosis differentiation rate is >80%. These rapidly mutating Y chromosomal STRs showed the differential ability for both unrelated males and paternal relatives. These markers which are part of RM Y STRs panel also showed high mutation rates. Different commercial companies have proposed different panels of Y Chromosomal STRs to supplement the currently used set of 17 Y-STRs for increased paternal lineage differentiation such as power plex 23 and Y ler Plus, but none of these showed increase discrimination capacities to differentiated paternal relatives as RM Y STRs have. Moreover, these 13 RM Y STRs give us 29%-100% paternal lineages differentiation in most of the populations and 26%-85% in endogamous populations where conventional Y STRs fails to differentiate or have 4% paternal lineages differentiation power. This high rate of differentiation or individualization using classic RM Y STRs is a great bene t to the eld of forensic investigative genetics. Results of this study are concordant with previous studies 6, 15,16,20,21 and have showed considerably increased discrimination power. We also merged this data with previous data (and based on 5176 pairs of men related by 1-20 meioses), overall differentiation rate was 38.79% while Y-ler differentiation was 9.24% based on 3864 pairs. Using father-son pair data from the present and previous studies, we also provide updated RM Y-STR mutation rates. However, further studies should be conducted on Pakistani populations, mainly in comparison with commercial kits to further improve the mutation rate information. This study contributes to globally expanding databases for the set of 13 RM Y-STRs.