Detection of new variant “ Off-ladder ” at the D 12 S 391 , D 19 S 433 and D 1 S 1656 loci and tri-allelic pattern at the D 16 S 539 locus in a 21 locus autosomal short tandem repeat database of 400 Iraqi Individuals

The objectives of the study included the detection of genetic variation of 21 autosomal short tandem repeats (STRs) loci from random unrelated individuals in the middle and South Arab people of Iraq, the Forensic efficiency parameters of the autosomal 21 genetic loci using power plex21® kit and to evaluate the importance of these loci for forensic genetic purposes and the possibility to use the new kit in routine practical work. FTA® Technology was utilized to extract DNA from blood collected on FTATM paper. Twenty one (21) STR loci including D3S1358, D13S317, PentaE, D16S539, D18S51, D2S1338, CSF1PO, Penta D, THO1, vWA, D21S11, D7S820, TPOX, D8S1179, FGA, D1S1656, D5S818, D6S1043, D12S391, D19S433 and Amelogenin were amplified by using power plex21® kit. PCR products were detected by genetic analyzer 3130xL then, the data processed and analyzed by PowerStatsV1.2 software. Several statistical parameters of genetic and forensic efficiencies based on allelic frequencies have been estimated. This includes the observed heterozygosity (Ho), expected heterozygosity (He), paternity index (PI), random match probability (RMP), power of discrimination (PD), chance of exclusion (CE), polymorphic information content (PIC) and P-value. The power of discrimination values for autosomal tested loci was from 75 to 96% therefore those loci can be safely used to establish a DNA-based database for Iraq population for identifical purpose. The high PIC values of the selected markers confirm their usefulness for genetic polymorphism studies and linkage mapping programs in human as well. The mean heterozygosity observed, expected and mean PIC values across the 20 loci were 0.79, 0.83 and 0.81, respectively indicating high gene diversity. A total of nine off-ladder alleles and one tri-allelic pattern were detected in this study.


INTRODUCTION
Microsatellites are a group of molecular markers chosen for a number of purposes including forensics individual identification and relatedness testing (Nakamura et al., 1987;Yamamoto et al., 1999).Microsatellites refer to DNA with varying numbers of short tandem repeats (Allor et al., 2005;Klintschar et al., 2005) between a unique sequence.DNA regions with repeat units that are 2 to 7 bp in length or most generally short tandem repeats (STRs) or simple sequence repeats (SSRs) are generally known as microsatellites (Ellegren, 2004).In the core repeated bases, long repeat units may contain several hundred to thousands (Butler and Hill, 2012;Silvia et al., 2009).DNA can be used to study human evolution using human genome analysis regions that are not subjected to selection pressure (Mats et al., 2007;Mohammed and Imad, 2013).Besides, information from DNA typing provides vital information in medico-legal with polymorphisms allowing for more biological studies (Walkinshaw et al., 1996).It has been found that microsatellites are evenly distributed in the genome on all chromosomes and all regions of the chromosome (Ensenberger et al., 2010).They can also be found inside gene coding regions, introns, and in the non-gene sequences.Most microsatellite loci are really small, ranging from a few to a few hundred repeats and this small size of microsatellite loci is important for PCRfacilitated genotyping.Basically, microsatellites containing a higher number of repeats are more polymorphic.
The number of repeats at a given locus can vary between individuals and there is a statistical probability that a given individual will have a set number of repeats at a particular STR locus (Moxon and Wills, 1999;Butler, 2005;Imad et al., 2014a;Muhanned et al., 2015).To process the results from each analysis, large database known as combined DNA index system (CODIS) has been set up.This database stores profiles from convicted offenders and unsolved casework.Similar databases have been set up in Europe, Japan, and other countries.The information in these databases can be used to detect and apprehend serial offenders by permitting rapid exchange of information between crime laboratories (Budowle et al., 2000;Ruitberg et al., 2001;Imad et al., 2014b).
Autosomal STRs locus information were evaluated and selected at The Cooperative Human Linkage Center.http://www.chlc.orgevaluates the genetic markers and the loci are selected from there.Additional STR loci, chromosomal location and repeat sequence for each core STR locus are provided in Tables 1 and 2 (Klintschar et al., 2005).According to the International Society of Forensic Genetics (ISFG) recommendation, the repeat sequence motif was defined so that the first 5'-nucleotides on the Gene Bank forward strand define the repeat motif used; therefore, the repeat motif for each STR marker is listed based on this.A significant fact is that STR allele sizes are measured relative to an internal size standard during electrophoresis.This depends on the DNA strand that is labeled using a dye that may have a different apparent measured size.The PowerPlex® 21 System is compatible with automated PCR instrument and with the ABI PRISM® 3100, 3100-Avant, 3130, 3130xl, 3500 and 3500xL Applied Biosystems Genetic Analyzers.In the United States, Europe and Asia, the PowerPlex® 21 System is used, and it increases the discriminatory power and data-sharing possibilities by incorporating informative loci.

Preparation of blood stain samples
Blood samples were randomly collected from 400 healthy unrelated males and females living in the middle and south of Iraq and sent to the forensic genetic laboratories in (DNA Typing of Medico-Legal Institute, Baghdad, Iraq) and the same samples were sent to Department of Biomedical Science, University Putra Malaysia, Selangor, Malaysia to complete other tests.

DNA extraction
DNA was extracted from all dried blood samples on FTA cards following the manufacturer's procedure as described in Whatman FTA Protocol BD01 (Mullen et al., 2009;Imad et al., 2014c).A 1.2 mm diameter disc was punched from each FTA card with a puncher and put directly in PCR tube.

PCR amplicon analysis (capillary electrophoresis)
The major application of capillary electrophoresis (CE) in forensic biology is in the detection and analysis of short tandem repeats (STRs).STR markers are preferred because of the powerful statistical analysis that is possible with these markers and the large databases that exist for convicted offenders' profiles.Other related applications include the analysis of haploid markers in the Y chromosome and in mitochondrial DNA (mtDNA) (Imad et al., 2014).Nonhuman DNA testing can also be performed depending on the circumstances of the case.The techniques involved include genotyping, DNA sequencing and mutation detection.The analysis of STR loci in DNA is the most common method for the between two or more unrelated determination of human identity and can *Corresponding author.E-mail: imad_dna@yahoo.com.Tel: 009647716150716.
Creative Commons Attribution License 4.0 Author(s) agree that this article remain permanently open access under the terms of the International License
Using the ABI Prism1 3730xl Genetic Analyzer, 16-capillary array system (Applied Biosystems, Foster City, CA, USA) following manufacturer's protocols, with POP-7™ Polymer and Data Collection Software, GeneMapper® V3.2 software (Applied Biosystems, Foster City, CA, USA) were done.By comparison of the size of a sample's alleles to size of alleles in allelic ladders for the same loci being tested in the sample, the STR genotyping was conducted.By comparison of the size of a sample's alleles to size of alleles in allelic ladders for the same loci being tested in the sample, the STR genotyping was conducted (Weir, 1996;Butler et al., 2004).

Statistical analysis for autosomal STR
The PowerStatsV1.2 (Promega, Madison, USA) was used to calculate the observed heterozygosity (Ho), power of discrimination (PD), probability of exclusion (PE) and polymorphism information content (PIC).Arlequin software program was used to conduct the exact test of population differentiation.In addition, Arlequin software program was used for the expected heterozygosity (He), and Hardy Weinberg Equilibrium (HWE).Where test results with P-values less than 0.05 were observed, the Bonferroni correction had to be applied to the data.The Bonferroni procedure (Weir, 1996) adjusted the rejection level for the smallest P-value at an overall level of α = 5 % to 0.05/x; where, x is equal to the number of tests conducted on the data.The Ho and He values were calculated by means of the same software program.
Following are the formulas used to compute various parameters for population data analyses: Expected heterozygosity Edwards et al. (1991) described the following formula for calculating an unbiased estimate of the expected heterozygosity: Where, n1, n2, …, nk are the allele counts of K alleles at a locus in a sample of n genes drawn from the population and pj is the allele frequency.

Match probability
The probability of a match at locus l PMl, was first described from genotype data (Fisher, 1951): Where, is the sample frequency of the ith genotype at locus l.
Where, the first part of this equation is for a sample of size Nl at locus l (Jones, 1972).

Power of discrimination
Brenner and Morris (1990) described the following formula for calculating the power of discrimination:

Polymorphism information content
The PIC was also calculated using marker allelic frequencies using the following equation: Where, n is the number of alleles and pi is the allele probability of the ith allele (Botstein et al., 1980).

Power of exclusion
Where, H = heterozygosity.

Paternity index
Where, Pi is the frequency of ith allele in a population of n samples; h = homozygosity and H= heterozygosity.

Allele frequency of common autosomal genetic loci
After the samples have been collected, DNA extracted and PCR amplified, they were genotyped for the 20 STR loci of interest.The genotyping information was then converted into allele frequencies by counting the number of times each allele was observed.Allele frequencies for each of the 20 STR loci in the Iraq population sample are shown in Tables 3 and 4. Since there are some alleles which were not sampled sufficiently and that an estimate of an allele frequency is uncertain if the allele is so rare that it is represented only once or a few times in a dataset, it is recommended that each allele be observed at least five times to be used in forensic calculations (Butler et al., 2009).The minimum allele frequency is 5/(2n); where, n is the number of individuals sampled and 2n is the number of chromosomes (as autosomes are in pairs due to inheritance of one chromosome from each parent).In the loci D5S818 (allele 13), the highest allele frequencies are found, and the lowest allele frequencies are at allele13 as seen in D3S1358 locus.D21S11 and D18S51 loci illustrate the largest number of different alleles.The following locations, the most common alleles at the 21 loci, were allele 13 for (D5S818, D8S1179 and D18S51), allele 10 for (D7S820 and Penta D), allele 16 for (D2S1338 and D2S1338), allele 11 for (TPOX, D16S539, PentaE, CSF1PO and D6S1043) loci, allele 12 for (D13S317 and D1S1656), allele 30.2 for D21S11 locus, allele 9 for THO1 locus, allele 13.2 for D21S11 locus, allele 6 for THO locus, alleles 17.3 and 16 for VWA locus, allele 30.2 for D19S433 locus, allele 17 for D3S1358 locus, allele 18 for D5S818 locus, allele 18 for D12S391 locus, allele 14.2 for D12S391 locus and allele 25 for FGA locus.The polymorphic nature of microsatellites (STR) makes them the markers of favorite in properties and genetic diversity studies (Ossmani et al., 2009;Chouery et al., 2010;Imad et al., 2014d).The best indicator of the genetic polymorphism within the sample is verified by the number of alleles and the expected heterozygosity found in the Iraq population.Basically the number of alleles is highly associated with the size of the sample.This is due to the presence of unique alleles in populations, which occur in low frequencies.The usefulness of the markers for genetic screening is verified by the number of alleles scored for each marker (Andreini et al., 2007).
The number of alleles and the expected heterozygosities detected in Iraq population are good indicators of the genetic polymorphism within the breed.Generally the number of alleles is highly dependent on the sample size because of the presence of unique alleles in populations, which occur in low frequencies and also because the number of observed alleles tends to increase with increases in population size.The number of alleles scored for each marker is an invaluable indicator of the future usefulness of the marker for genetic screening.Finding the same number of alleles for certain different loci in various populations (for example, Iran, Syrian, Emirates, Qatar and Egyptian populations) may indicate common ancestries (Reyhaneh and Sadeq, 2009;Alshamali et al., 2003;Ana et al., 2003;Clotilde et al., 2007).The frequency and the number of alleles, however, may be an indication for the degree of inbreeding within each population and thus reflects the homogeneity of the population.The analysis of STR polymorphisms by PCR-based method offers certain advantages over RFLP typing.In recent years, short tandem repeat (STR) systems have gained importance in forensic analysis of biological specimens as well as in paternity testing, as an alternative to the use of restriction fragment length polymorphism (RFLP) analysis (Alshamali et al., 2003;Ana et al., 2003;Clotilde et al., 2007;Hammond et al., 1994).The analysis of STR polymorphisms by PCR-based method offers certain advantages over RFLP typing: (1) STR loci can be typed with a high degree of specificity and sensitivity in a short time period, (2) these loci can be successfully amplified from a limited amount of DNA even if it is degraded, and (3) typing of multiple loci can be accomplished in a single multiplex reaction (Hochmeister et al., 1991;Lins et al., 1996).

The amelogenin locus
The Amelogenin locus that occurs on both the X and Y chromosomes and enables sex typing (Sullivan et al., 1993) was also located within the reference human genome sequence.AMELX is located on the X chromosome at 10.676 Mb.AMELY is located on the Y chromosome at 6.441 Mb.Amplification of Amelogenin generates different length products from the X and Y- chromosomes.Figure 1 shows the possible peaks for both genotypes X/X and X/Y represented as one and two peaks, respectively, in the GeneMapper electropherogram.

New additional autosomal genetic loci
To improve results with challenging DNA samples, a set of 20 autosomal STR loci was characterized in our

Allele
laboratory at NIST (Hill et al., 2009).The efforts of the Human Genome Project have increased knowledge regarding the human genome, and hence there are many more STR loci available now than there were 10 years ago.In fact, more than 20,000 tetranucleotide STR loci have been characterized in the human genome (Collins  et al., 2003) and there may be more than a million STR loci present depending on how they are counted (Ellegren, 2004).STR sequences account for approximately 3% of the total human genome (Landerand Linton, 2001).D12S391 is one of new nucleotide repeat STR loci and eight alleles were found in this study.As shown in Figure 2, the allele 18 was most frequent 0.359.

Allele frequency of Penta D genetic loci
Penta D is a pentanucleotide repeat found on chromosome 21 about 25 Mb from D21S11.Alleles ranging from 1.1 to 19 repeats have been observed although some of the shorter alleles are likely due to flanking region deletions (Kline et al., 2011).Penta D is present in the PowerPlex® 16 and PowerPlex® 18D kits.Eleven (11) alleles were found in this study, and as shown in Figure 3, the allele 10 was most frequent 0.398.

Allele frequency of penta E genetic loci
Penta E is a pentanucleotide repeat with very low stutterproduct formation that is located on the long arm of

Allele frequency of D6S1043 genetic loci
D6S1043 is a compound tetranucleotide repeat with alleles ranging from 8 to 25 AGAT or AGAC repeats.Some x.2 and x.3 alleles have been reported in population studies.D6S1043 is part of the Sinofiler™ kit and has been used to date almost exclusively in Chinese and other Asian population studies.D6S1043 is located less than 4 Mb from SE33 on the long arm of chromosome 6.Eleven (11) alleles were found in this study, and as shown in Figure 5, the allele 11 was most frequent 0.238.

Allele frequency of D1S1656 genetic loci
D1S1656 is a tetranucleotide repeat found on the long arm of chromosome 1 with alleles ranging from 8 to 20.3 repeats.The x.3 alleles arise from a TGA insertion typically after four full TAGA repeats.It is part of the

Forensic efficiency parameters
Statistical analysis is used to interpret DNA results for genetic identity.In order to determine the significance of a match, it is necessary to support DNA typing results with statistical analysis.These analyses assign a value to the results obtained and enable easier resolution of forensic or paternity cases.Across all loci, the values for the matching probability, power of discrimination, power (chance) of exclusion, polymorphism information content and typical paternity index for the 20 STR loci of the Iraq population were determined and are indicated in Figures 7 to 14.

Observed heterozygosity and expected heterozygosity
The two common ways to report heterozygosity are observed and expected heterozygosities.Observed heterozygosity is calculated by dividing the number of heterozygote individuals at a locus by the total number of all individuals at that locus and describes the heterozygosity actually observed in the members of the sampling group.Expected heterozygosity is calculated as 1 minus the homozygosity (the sum of squares of all allele frequencies at a locus) and represents the number of heterozygotes that would be expected under HWE based on the observed allele frequencies in the sampling group.A higher heterozygosity means that more allele diversity exists, and therefore there is less chance of a random sample matching.Observed heterozygosity and expected heterozygosity all over the 20 loci are presented in Figures 7 and 8, and the observed heterozygosity oscillated between studied populations as illustrated in Table 5.The observed heterozygosity in a population relies on the number and the frequency of alleles of each locus.Moreover, the distribution of genotypes in a population sample may deviate from HWE expectation in a number of ways.These include the presence of an excess of homozygotes and a corresponding lack of heterozygotes or an excess (deficiency) of one or more classes of heterozygotes or a combination of those states.There are populations with low heterozygosity, lower than 65% in most tested loci.These populations are small, closed, inbred by cultural or geographical factors; one of those populations is the Qatari population where the levels of observed heterozygosity (Ho) oscillated, between 0.339 for D19S433 and 0.839 for D2S1338.Interestingly, Ho is lower than expected heterozygosity (He) in almost all the analyzed loci, with the unique exception of D5S818 locus.This fact was particularly conspicuous for the loci that deviated from the HWE expectations; the departures from HWE expectations detected in the Qatari population seem to be the result of excess of homozygotes over heterozygotes, which is likely to be the consequence of the high consanguinity rates reported for this population, which is 46% (Ana et al., 2006).

Paternity index
The paternity index (PI) compares the likelihood that a genetic marker (allele) that the alleged father (AF) passed to the child to the probability that a randomly selected unrelated man of similar ethnic background could pass the allele to the child.The potential of a randomly selected man to pass the obligate gene is determined by using a database, which lists the frequency distribution of individual alleles within a given genetic system.Combined paternity index is an odd ratio that indicates how many times more likely it is that the alleged father is the biological father than a randomly selected unrelated man of similar ethnic background.The paternity index was high for all STR analyzed; it ranged from 2.651 (TPOX) to 2.864 (D21S11) (Figure 9).

Random match probability
The match probability is the probability for a random match between two unrelated individuals drawn from the same population.It is the sum of the frequency squared of each genotype its that ranged from 0.011 to 0.168 (Figure 10).
A random match probability, is the probability that a person sampled randomly from the population, would have a particular STR profile.If assumptions can be made that (1) alleles within a locus are inherited independently and (2) that STR loci migrate independently of one another during meiosis, then it is possible to multiply the specific expected genotype frequencies (based on allele frequencies) for all the loci together to obtain the overall random match probability.This combination of all the loci match probabilities by multiplication is termed the product rule.The strength of a genetic match between a questioned sample and a known sample (rarity of a particular DNA profile) relies on the discriminative power of the product rule.Thus the assumptions of independence for alleles within a locus and independence among migrating loci are central to determining the strength of a match (Butler et al., 2005).

Power of discrimination
Power of discrimination is defined as the probability that   that the previously mentioned loci have a similar degree of polymorphism (Ana et al., 2006;Reyhaneh and Sadeq, 2009).The polymorphic nature of microsatellites makes them the markers of choice in characterization and genetic diversity studies.The high PIC values of the selected markers confirm their usefulness for genetic polymorphism studies and linkage mapping programs in human as well.The mean heterozygosity observed, expected and mean PIC values across the 20 loci were 0.77, 0.81 and 0.78, respectively, indicating high gene diversity.

P-value: probability value of Chi-square test for Hardy Weinberg equilibrium
Checking for HWE is performed by taking the observed allele frequencies and calculating the expected genotype frequencies based on the allele frequencies.If the observed genotype frequencies are close to the expected genotype frequencies calculated from the observed allele frequencies, then the population is in Hardy-Weinberg equilibrium and allele combinations are likely to be independent of one another.The results obtained from the tests for HWE is shown in Figure 14.The null hypothesis states that all the STR loci tested are in HWE and any deviation from HWE expectations is due to sampling error (Hill et al., 2009).The alpha value was set at 0.05.The null hypothesis was rejected if the computed p values were below the alpha value of 0.05.Therefore having a p value above 0.05 to show that the STR alleles do not differ significantly from HWE and does not imply that the samples are in HWE.Butler (2005) states that -if a p-value of < 0.05 is observed with a set of alleles measured at a particular STR locus, it does not mean that a laboratory should avoid using this data because it ‗failed' a test for Hardy-Weinberg equilibrium‖

Most common genotype frequencies
Another measure to reflect the usefulness of a particular set of DNA markers is to examine the frequencies of the most common genotypes, which would therefore be the least powerful in terms of being able to differentiate between two unrelated individuals (Edwards et al., 1991).The theoretically most common type can be calculated by considering a sample type that is heterozygous at all loci possessing the two most common alleles at each locus (Lander and Linton, 2001;Imad et al., 2014e).In Table 9, frequencies from the two most common alleles at each of the twenty loci were used to estimate a theoretical most common STR profile.A number of the newly available STR loci, such as D1S1659 and D6S1043, provide a better probability of identity than widely used loci such as D8S1179 and D2S1338.The -most common genotypes frequency‖ results column can also be a useful metric to locus performance.For example, a comparison of D1S1659 and THO1 is instructive.These two loci both have 44 most common allele frequency, yet D1S1659 has 0.2153 common genotypes frequency while THO1 only has 0.0851.The greater number of genotypes formed with the different combinations of alleles in D1S1659 leads to better probability of identity (0.050 vs. 0.011) values.Furthermore, additional genotype combinations mean that D1S1659 will likely be more useful than THO1 for detecting contributors in DNA mixtures.

New Tri-allelic patterns have been observed for many of the core STR loci
Tri-allelic patterns have been observed for many of the core STR loci and recorded on the NIST STRBase website.Clayton et al. (2004) described possible reasons for tri-allelic patterns, which can occur as an imbalance in amounts between the three alleles (type 1) or equal amounts of all three alleles (type 2).A type 1 tri-allelic pattern imbalance is typically a situation where the sum of the peak heights for two of the alleles is approximately equivalent to the third allele (Clayton et al., 2004).It is interesting to note that TPOX, which occurs closest to the tip of a chromosome, has the highest number of observed tri-allelic patterns-most of which are type 2 with equal intensity alleles.Thus, it is possible that this section of chromosome 2 is more likely to be duplicated in some individuals for telomere maintenance to keep the end of the chromosome intact (Chakhparonian and Wellinger, 2003;Louis and Vershinin, 2005).One triallelic pattern D16S539 observed in this study is shown in Table 11 and Figure 18.A number of explanations for these three-banded allele patterns have been suggested: (a) a  None reported yet in STRBase --D19S433 None reported yet in STRBase --SE33 None reported yet in STRBase --D6S1043 None reported yet in STRBase --D1S1656 None reported yet in STRBase -- (1999) individuals, only 19 tri-allelic patterns (18 at TPOX and 1 at CSF1PO) were observed (0.19% per locus).In this study, one tri-allelic pattern was observed at the TPOX locus of sample M-200.This sample was genotyped twice on the ABI 310 to verify the result.The TPOX trialleles are 8, 10, and 11.This sample was neither included in the STR allele frequencies study nor the HWE database validation.The presence of an extra peak at one locus out of all of the 15 loci tested indicates that this sample is not a mixture.Rather, the extra peak is a real reproducible artifact of the sample.The sample was retyped and the same tri-allelic pattern was obtained.The fact that the peak heights of all three alleles are similar suggests that the alleles are probably present in equal copy number (Egyed et al., 2000).Tri-allelic patterns appear to be common at the TPOX locus.A search of the STRBase website (Butler, 2006) as of May 2006 reveals 14 different TPOX tri-allelic patterns.

Conclusion
The conclusions that have been drawn from the present study include: the results of the current study indicate these new autosomal STR Loci useful For DNA typing markers in Iraq can be used for establishment of a DNA database that will be beneficial for the population in terms of resolving social and moral disputes and will contribute to improvements in the justice system.The match probability is the probability for a random match between two unrelated individuals drawn from the same population; it ranged from 0.011 to 0.168.The mean heterozygosity observed, expected and mean PIC values across the 20 loci were 0.77, 0.81 and 0.78, respectively, indicating high gene diversity.A total of nine off-ladder alleles were detected in this study; four of them are located at the D1S1656 locus, three are located in D12S391, one occurred at the D19S433 locus and one discovered outside the range.One triallelic pattern observed in this study was 9/12/13 pattern.

Figure 1 .
Figure 1.An electropherogram showing Amelogenin locus and its two possible genotypes obtained with PowerPlex®21 amplification and GeneMapper version 4.0.

Figure 3 .
Figure 3. Allele frequency of Penta D genetic locus.

Figure 4 .
Figure 4. Allele frequency of Penta E genetic locus.

Table 1 .
Comparison of STR loci present in kits used in the United States.
in length.A number of same-size, differentsequence alleles have been identified through sequence analysis.It is part of the extended European Standard Set and is present in NGM™ and NGM SElect™ kits from Applied Biosystems, the PowerPlex® ESI and ESX Systems from Promega, and ESSplex and ESSplex SE kits from Qiagen.

Table 5 .
The observed heterozygosity in different populations.

Table 9 .
Most common genotype frequencies based on the two most common alleles found in a Iraq population.

Table 10 .
New variant or -off-ladder‖ discovered in this study and comparison with alleles reported in STRBase.

Table 11 .
A total tri-allelic pattern observed in our study and reported on STRBase.