Introduction

The experiments performed by Gregor Johnn Mendel during 1860’s on the garden pea plant laid a foundation of understanding the basic principles of genetics, commonly known as Medalian genetics. Based on simple pedigree analysis it was made clear that the characters are inherited from parents to their off springs. During his studies, Mendel postulated some basic principles like Dominance, Segregation and independent assortment of characters. In the modern era, the study of inheritance performed at the molecular level is based on some seemingly esoteric concepts, which of course need to be resolved at the outset so that the concept is understood as a whole. The human genome is composed of millions of nucleotides and only a small portion of it is utilized as a tool in differentiating individuals from each other, first done by Sir Aliec Jaffrey in 1984 (Jeffreys et al., 1985). The technique however remained less appreciated and was mostly employed in demystifying the genetic disorders due to malfunction in some of the genes in the close proximity of the repetitive sequences. Soon after the value of genetic diversity among these repetitive sequences was recognized, more and more research started focusing on recognizing the potential hotspots for better variability and stability. The invention of Polymerase Chain Reaction in 1985 by Kary B. Mullis proved a milestone as the technique was based on a powerful concept of generating multiple copied from the desired segments of DNA. The invention was unequivocally a major breakthrough to defeat the constraints of low template DNA, met mostly in forensic samples. The quest of designing better and more reliable hotspots having better polymorphic characters and greater stability, more genetic markers were added in order to increase the discrimination power. Consequently, a huge number of markers have already found there place in various multiplex kits. Even the markers from same chromosome, having impact of linkage are being selected which of course are going to actually disturb the calculated value of evidence, if the linkage factor is ignored. In this chapter we shall try to evaluate the effect of two important concepts –Linkage and Mutations on STR typing and interpretation of results. However, before going in to the discussing precisely, there are certain basic concepts which need to be understood first especially for the beginners.

A diploid cell undergoes a series of sequential modifications before giving rise to a germ cell. The most important one of these changes takes place during prophase-I of meioses where the homologous chromosomes align to undergo exchange of segments during recombination. In human cells (whether male or female), the two homologs align adjacent to each other and take part in the formation of chiasmata during crossing over except sex chromosomes in human males, that will be discussed later. The autosomal (AS) markers have two alleles at one locus, of which one is from father (paternal) and the other one is from mother (maternal). After a successful recombination event, the chromosomes are separated towards opposite poles of the cell during anaphase-I. In AS chromosomes recombination serves as a driving factor towards polymorphism bringing in more and more combinations. Same is the case with sex chromosomes in human females (XX) which also undergo recombination leading to formation of recombinants. However, the Y-chromosomes, in human males do not have a counterpart to pair with and hence there is no recombination between XY. The only X chromosome in male child which has been contributed by the mother is a recombinant one while in case of a female child, one X chromosome contributed by the mother is a recombinant whereas the other one contributed by the father is a non-recombinant.

The genetic information stored in the DNA is distributed among 46 chromosomes having 44 autosomes and 02 sex-determining chromosomes, in humans. Only about 0.1 percent of this genetic information is found to be polymorphic, of which Short Tandem Repeats (STRs) are the most targeted hotspots in the contemporary forensic DNA typing. Short Tandem Repeats (STRs), are simple sequence repeats having a repetitive unit of 1-6 bp (Tautz, 1993). These repetitive sequences are scattered in the genome of both Prokaryotes as well as Eukaryotes including human. Most of these STR sequences are found in non-coding parts and only a small portion in coding part (Ellegren, 2000). These STRs are commonly found to be A-rich and their concentration is mostly on chromosome-19. On an average, in humans there is one STR sequence per 2000bp (www.genome.gov) There is a tendency for selection of these STR markers based on their polymorphic potential coupled with greater stability and reduced allele spread, as far as their role in forensics is concerned. A marker with greater discriminating power, low mutation rate and reduced length based alleles at a marker are the markers of choice. The reduced length based allele spread is desired because when there is less size difference between two heterozygous alleles, there are less chances of ambiguity in mixture analysis (Walsh et al., 1992). A large number of autosomal STR markers have been selected for this purpose over the last three decades; however there are only a limited number of these markers which can actually pour in greater variability as far as the STR profile is concerned. The second generation multiplex (SGM) having TH01, VWA, FGA, D8S1179, D18S51 and D21S11 markers provided a match probability of about one in five million. Introduction of other highly polymorphic markers like SE33, D2S1338, D18S51, etc. have added a great deal to the level of polymorphism. (Table 1.1) gives the probability of identity, which is the pair wise comparison of genotypes at a locus. Markers having lower probability of identity are more variable and hence more valuable.

Linkage

A concept given by Morgan, states that when markers tend to remain together during their inheritance they are said to be linked. Markers that are physically near to each other are unlikely to be separated by the recombination event during crossing over, thus are said to be linked. When linkage plays a role, the results will differ from, as predicted by the law of independent assortment. The distance between any two loci is measured in centimorgans (cM) which is defined as the distance between markers in which there is one percent (1%) chance of one marker being separated from other as a result of recombination in one generation. 1 cM is equal to recombination of one percent. Markers with recombination frequency of less than 50% are believed to be present on same chromosome and are considered as linked, whereas a recombination frequency of 50% depicts that the markers are present on non-homologous chromosomes or, are far apart on the same chromosome and hence considered as unlinked. Therefore, recombination frequency gives an idea of the level of genetic linkage between the markers, and can be calculated as;

Recombination frequency (θ) = (Number of recombinant progeny)/(Total number of progeny) X 100

In STR typing, especially while dealing with close relationship studies the impact of linkage is obvious but to calculate exact likelihood, we must be able to assess the level of impact also. This subject will be dealt with, in detail in the proceeding discussions.

Linkage disequilibrium (LD)

Also known as allelic association, is another phenomenon that needs to be considered while calculating LR. LD is said to exist when alleles of different loci tend to remain together in a population study more often than expected. It reflects recombination in all generations since the evolution of the allele under study. In other words when the probability of the occurrence of two alleles together (P AB) is not equal to their individual probabilities (PA PB), believing the association as completely random, LD is said to exist. LD is influenced by the factors like demographic history, Admixtures, etc. (Ardlie et al., 2002; Laan and Pääbo, 1997; Zavattari et al., 2000).

Impact of linkage on STR typing

No doubt forensic evidence at the scene of crime is a potent and robust source of information linking crime with criminal, but its weight as an evidence is determined by the assessment of associated factors. In order to bring in more and more variability between any two DNA profiles, addition of more STR markers is made based on basic characters of marker designing. Although it is preferable to use markers from different chromosomes, however modern and emerging multiplex kits (Table 1.2) are using markers close to each other on same chromosomes which intern raises the question of allelic association. In forensic estimation of relatedness among closely related individuals, linkage becomes important in calculating probability of passing an allele from father to child. If the LR calculated involves H1, suspect as donor and H2 as an unrelated person to the suspect, as donor then there is no impact of linkage on LR calculations. However, if the alternative hypothesis accepts a close relative of the suspect as the donor, the linkage will undoubtedly have to be taken in to consideration while calculating LR (Buckleton and Chris, 2006).

LR=H1/H2

Same is the case with impact of linkage on paternity Trio/Duo cases. When the alternative hypothesis considers an unrelated person as the alleged father, linkage will have no impact on calculations. However, when the alternative hypothesis involves any close relative as the alleged father, then the linkage will come in to play and needs to be considered while interpretation of results. (Table 1.2) gives a list of AS STR markers which are considered in forensic multiplexes despite the fact that there are more than one markers selected from same chromosome (Gill et al., 2012).

As we know the distance between any two markers in a genetic map has a direct impact on recombination frequency and thus on linkage. Therefore, the impact of linkage increases as the distance between the markers decreases.

An alternative approach as recommended by Budowle et al. (Budowle et al., 2011) for kinship analysis while having impact of linkage, is to either incorporate the recombination rate for generating maximum likelihood estimates of haplotype frequencies for two loci or opt for the most informative loci, out of the two in the calculations.

Table No. 1.1: Rank wise list of AS STR markers chosen for their variability (Butler et al., 2009; strbase.nist.gov).

Marker Rank Probability of Identity Number of alleles

SE33 1 0.0066 120

D2S1338 2 0.0220 22

D1S1656 3 0.0224 16

D18S51 4 0.0258 55

D12S391 5 0.0271 18

FGA 6 0.0308 98

D6S1043 7 0.0321 -

D21S11 8 0.0403 92

D8S1179 9 0.0558 19

D19S433 10 0.0559 30

vWA 11 0.0611 29

D7S820 12 0.0726 32

D16S539 13 0.0749 21

D13S317 14 0.0765 20

TH01 15 0.0766 21

D2S441 16 0.0841 11

D10S1248 17 0.0845 12

D3S1358 18 0.0915 27

D22S1045 19 0.0921 12

CSF1PO 20 0.1054 23

D5S818 21 0.1104 16

TPOX 22 0.1358 17

Table No. 1.2: List of AS STR markers chosen, despite an inter-marker distance Constraint, for forensic use (Gill et al., 2012)

Location on chromosome Locus pair Distance in between

5q 23.2 D5S818 25 cM (6)

5q 33.1 CSF1PO

2q 35 D2S1338 50cM (6)

2q25.3 TPOX

21q21.1 D21S11 50cM (7)

21q33.1 Penta D

2q35 D2S1338 Located on different arms

2p14 D2S441

12p13.31 vWA 12cM (8)

12p12 D12S391

Linkage in X STRs

Unlike autosomes, the X-chromosomes in males do not occur in pairs and thus cannot form homologs. Here we have only one X-chromosome and one Y-chromosome. In case of human females, there are two X-chromosomes forming a homologous pair which undergoes recombination during meioses. X-chromosomal STR markers have gained a special scope in forensic science especially in kinship cases where the alleged father cannot be traced and the child is a female (Gusmão et al., 2012; Szibor et al., 2007). Although while dealing with standard paternity cases like duos/trios, it is highly desirable that AS STRs should be preferred. However, there are certain conditions when the use of X STRs becomes essential, either as additional or exclusive. For instance, in case of few genetic inconsistencies between the alleged father and the daughter, the inconsistencies are explained as the result of mutation or the alternative hypotheses is considered as the alleged father being genetically close to biological father (Gomes et al., 2012). If the biological father is the father of the alleged father, with different mothers (Fig. 1.1) or if the biological father is the son of the alleged father, with different mothers (Fig. 1.2), they are not going to share X STR alleles, identical by descent (IDB).

Figure No. 1.1: “A” the biological father is the father of the alleged father “B”. “C” is the daughter in issue. “A” and “B” do not share the x-chromosome

Figure No. 1.2: “A” being the biological father of “C” and “A”,the alleged father. “A” and “B” do not share the X-markers

X STR typing is also important when there are no direct sample sources for matching the profile in a deficiency case. As discussed earlier, the sisters inherit partially matching haplotypes from the mother. In these cases when the putative father is not available, the mother of the putative father (grandmother of female child) is more informative and Mean Exclusion Chance (MEC) can be generated (Krüger et al., 1968). which is a chance of excluding a man who is not the father by an inconsistency in at least three loci, considering mutations.

Here it must be remembered that these complications arise for AS STRs while dealing with close genetic relations. For example, in complete siblings the probability of having same allelic combination at any marker is ¼ (0.25). This means there are chances that the fifth child born from the same parents is expected to have the same combination of alleles as any one of the other four. With the increase in the number of markers used, of course the probability of repetition of same combination will go on decreasing as a result the AS STR profile gives a better conclusion and the authenticity of the profile generated increases (Fig.1.3). Nonetheless, due to low variation in between closely related individuals like siblings, it is preferable to opt for X STR typing when the above discussed scenarios are faced.

Figure No. 1.3: Using any marker in a full sibling case, a probability of 0.25 exists, that offspring beyond the four given in the figure, will repeat the genotype of any of the four given

Using X-STR markers is more authentic than AS STRs in such cases because of their inheritance pattern. The males contain one X-chromosome (XY) and hence the haplotype is transmitted from biological father to biological daughter without undergoing any recombination. However, in case of females there are two X-chromosomes (XX) and thus the recombination is possible. Therefore, while considering X STR markers for relationship tests two important factors need to be considered while interpreting the results. These are linkage (inheritance of closely related markers together with higher probability than physically separated markers) and linkage disequilibrium (At a population level, occurrence of alleles of different loci together, more or less frequently than expected by chance).

We know that recombination is a direct measure of linkage between any two markers. A recombination frequency of less than 50% (50 cM) is considered as the indication of linkage between any two markers while a 50% recombination frequency is an indication that the two markers are either present on different chromosomes or are far enough on the same chromosome to assort independently. While dealing with X-chromosome with an approximate length of about 155 Mb (180 cM) (International HapMap Consortium et al., 2007). only 3-4 STRs can be considered which are at least separated by a minimum distance of 50 cM, required to ensure their independent assortment. However, additional markers are introduced (Szibor et al., 2007). which can greatly diminish the results, if linkage factor is not taken in to consideration. It is therefore recommended that while dealing with X STRs in kinship analysis the guidelines put forth by DNA commission for International Society for Forensic Genetics (ISFG) must be adhered to, before interpreting the results. The recommendations are summarized as under (Tillmar et al., 2017).

Guidelines put forth by DNA commission for International Society for Forensic Genetics (ISFG)

Mutations and their impact on the interpretation of STR results

Although STRs are wide spread in both prokaryotes as well as in eukaryotes, they are mostly considered as junk DNA having no biological function. However, some recent advancements have revealed that in fact these STRs play an important role in many organisms (Escher et al., 2000; Gebhardt et al., 1999; Meloni et al., 1998). As discussed earlier, an STR marker is selected considering a basic set of characters and a novel polymorphic STR marker is the one with lowest mutation rate. However, this criteria is not immune everywhere and in case of Y STR study, the role of mutations gains more focus and plays an important role in various cases. The normal DNA sequences show a mutation rate of about 10-9 per generation as compared to the mutation rate of STRs, 10-6-10-2, which is several orders greater than the normal DNA sequence mutation rate (Ellegren, 2000). These mutations and more precisely, the mutation rates play an important role as far as the interpretation of results of a DNA profile is concerned. However, before going in to this discussion there are certain facts based on research, which should be taken in to account while framing the results.

• Dinucleotide repeats are found to be showing greater mutation rates as compared to tetranucleotide repeats (Chakraborty et al., 1997).

• Certain sequences like GC content have strong role in determining rate of STR mutations (Gymrek et al., 2017).

• Mutations having negative effect on the fitness are going to be eliminated from the population and recombination has an effect on doing away with these mutations through recombination repair. Therefore, it becomes evident that these recombination events may have an impact on the amount of mutations that are carried on downward and thus on the interpretation of results especially in paternal biogeographic ancestry. For example, Y-chromosomal markers do not undergo recombination and hence mutations in these markers will get transmitted in to the offsprings 100 %. This feature of Y-chromosomes makes it fit for patrilineal studies and tracing of ancestry (Kayser, 2017).

STR mutation rate in case of AS, and Y-chromosomes falls in between 1-5x10-3 ((Dupuy et al., 2004; Decker et al., 2008). and similar results are produced for X-chromosome STRs, however this needs extensive studies covering different populations to generate a concrete database. Study of mutations and their impact thereof, in STRs is probably the field of future research. (Table 1.3) gives the rate of mutation at the commonly used AS STR markers with TPOX and TH01 showing lowest rate while as the most polymorphic markers like FGA and SE33 showing the highest rate. (Table 1.4) gives the mutation rates at 15 X-STR markers.

At present, there are three possible mechanisms, which have been proposed for STR mutations. Although believed to have least effect, unequal crossover during meioses is considered one of the factors (Huang et al., 2002). Similarly generation of A-rich STR’s through 3ʹ extension of retrotranscripts by a mechanism known as retrotransposition (Nadir et al., 1996). However, the most widely accepted mechanism for STR mutations is “Strand Slippage Model”, proposed by Kornberg et al in 1964 (Kornberg et al., 1964).

The DNA polymerase during replication uses the information of the template segment to generate a complementary sequence. Since in STRs, each repeat unit is the same, e.g. (GATA) x the DNA polymerase cannot differentiate between the succeeding units and thus there are ample chances that some repeat units may be skipped. Sometimes the DNA polymerase and the daughter strand complex temporarily dissociate and due to the same repeat unit ahead, the pairing resumes and annealing occurs normally, as a result the left out segment forms a loop. Now, if this loop formation occurs on the newly synthesized strand during replication, the synthesized strand (daughter strand) will be, the number of repeat units which were skipped, longer than the template strand and vice versa. (Fig.1.4 A, B). It is however believed that there is difference in the apparent rate of strand slippage and effective mutations (Schlötterer and Tautz, 1992). which is believed to be done by the strand specific mismatch repair system, reducing the effective mutation rate by about 100-1000 fold (Strand et al., 1993).

Figure No. 1.4: Mechanism of Strand Slippage (A) the formation of loop occurs in the daughter strand and hence it will be one repeat unit more as compared to the template. (B) formation of loop occurs in the template strand and hence the new strand will be one repeat unit short as compared to template strand.

Although AS STRs are the markers of choice due to their high polymorphism, there are certain conditions under which the study of Y STRs becomes essential. For instance in case of multisource samples (as in gang rape), the concentration of source sample does not have uniform contribution from all the contributors. It is likely that the minor contributor will go scot free, using AS STRs. Similarly during rape, the slide prepared from the vaginal swab will no doubt contain the male contribution but, it is minor as compared to female background contribution. As a result there is serious competition for primer binding during PCR run with least chances of it (minor contribution) being primed. In order to do away with this, Y STR profiling becomes essential, first employed in 1992 ) (Roewer and Epplen, 1992) and considered three times more suitable in identifying male contributors as compared to AS STRs (Purps et al., 2015). As discussed in the introduction, the human females do not possess the Y-chromosomes, thus the 50% specificity stands achieved. The human males normally possess one X-chromosome and one Y-chromosome and the Y-chromosome has no homolog to pair with and thus the markers are transferred from father to the male child without undergoing recombination. When there is no recombination, there is no variation and the only source of variation in Y STRs is therefore the effective mutation rate. At present up to 27 markers are utilized in commercial kits for forensic use (Yfiler Plus by Thermo Fisher Scientific) (Gopinath et al., 2016). One of the striking features of Y STRs, which makes them essentially important to take up different assignments in forensic case work, is that different Y STR markers have different mutation rates. By using standard STRs with low-medium mutation rates (one or few mutations per 1000 per locus) (Goedbloed et al., 2009), paternity cases can be solved where the putative father is not available to provide a direct sample for comparison with the male child. In this case any male relative of the unavailable putative father can be used as a source of sample providing Y haplotype similar to putative father. It is worth mentioning here that the rapidly mutating Y STR (RM Y STR) cannot be utilized for paternity and kinship tests because the higher mutation rate in these RM Y STRs will sometimes confuse with inconsistencies at specific loci and intern trouble the probability calculations. However, these RM Y STRs have an important role to play where the target is to work out the individual identification when there are large number of male relatives traced from long patrilineal tree. It becomes very confusing when there is no difference in Y STR haplotype in a huge number of male individuals sharing same Y STR profiles. Therefore in these cases RM Y STRs, having a mutation rate in the order of 10-2 (Ballantyne et al., 2010) (Table 1.5) are very informative.

Y STRs also play an important role in studying paternal bio-geographic history. In absence of recombination in Y STR markers, when a mutation has occurred it cannot be removed from the gene pool, producing differences between the individuals of different geographic regions through factors like genetic drift. The impact of mutations and the selection of markers, from a variety of Y STRs available, becomes an essential step as far their use in forensic science is concerned. However, there are some specific guidelines provided by ISFG which need to be taken in to consideration while reporting a Y STR test. Given that different Y STR markers are transferred together in to, offsprings (single haplotype), it is not desired to multiply the individual allelic frequencies rather, the haplotype frequency is assessed against a relevant population in fairly large database. Similarly, mutation rates must also be taken in to consideration especially while dealing with kinship cases (Gill et al., 2001) (Butler et al., 2009). The profile of male staff members working in the laboratory should be recorded against Y STRs employing kits meant for general use. If increased PCR cycles are used the contamination factor must also be given due weightage (Gill et al., 2001).

Table No. 1.3: Gives an idea about the mutation rates at commonly used AS STR markers in the course of paternity testing (strbase.nist.gov).

STR Marker Total Number Of Mutations Mutation Rate (%)

TPOX 100/857481 0.01

TH01 100/779554 0.01

D7S820 1089/1085305 0.10

D5S818 1259/1107339 0.11

D16S539 1041/962239 0.11

D119S433 187/173490 0.11

D3S1358 1152/964288 0.12

D2S1338 262/225140 0.12

D8S1179 1239/899837 0.14

D13S317 1558/1103282 0.14

Penta D 57/41202 0.14

CSF1PO 1487/947425 0.16

Penta E 163/100030 0.16

vWA 2480/1437945 0.17

D21S11 1816/962096 0.19

D18S51 1746/790342 0.22

FGA 3125/1101006 0.28

SE33 330/51940 0.64

Table No. 1.4: Gives an idea about the mutation rates at 15 X-STR markers (Diegoli et al., 2014)

Marker Mutations Meioses Rate of Mutation (x10 -3)

GATA165B12 0 2333 0-1.6

DXS7130 0 1375 0-2.7

DXS10147 0 1429 0-2.6

DXS6795 0 1375 0-2.7

DXS101 1 3909 0.01-1.42

GATA172D05 1 2251 0.01-2.5

GATA31E08 1 2502 0.01-2.2

DXS7423 2 4890 0.05-1.5

DXS7424 2 3180 0.08-2.3

DXS6803 3 2390 0.26-3.7

DXS9902 4 1833 0.59-5.6

DXS6789 5 4853 0.33-2.4

DXS8378 8 5257 0.66-3.0

HPRTB 8 5905 0.59-2.7

DXS7132 15 6340 1.3-3.9

Table No. 1.5: Mutation rate at different Y STR markers used in different kits with corresponding mutation rates and number of RM YSTRs (Gymrek et al., 2017)

Conclusion

Over the years, more focus is laid on considering different potent factors which have strong correlation with STR typing and a viewed impact on the interpretation of results. Therefore, it is desired that the research in vogue regarding factors like linkage, linkage disequilibrium and mutations and their impact on the interpretation of results considering different scenarios faced in the forensic field, need a special focus. While dealing with linkage and linkage disequilibrium it is always desired to use computer generated programs for calculating LR to make them error free, especially in case of X STR analysis. In fact there are many software available but most of them assume markers as unlinked which intern makes the interpretation less reliable. However, software like Merlin include linkage and LD in the calculations but leave out the effect of mutations during LR calculations. FamLinkX is one such advancement which includes the impact of mutations also while dealing with X STR markers. Designing kits for X STR markers is one good step forward. Investigator Argus X-12 PCR amplification kit by Qiagen, uses twelve markers simultaneously and amelogenin for gender identification, can be helpful in various cases. While in case of Y STR markers it is always desired to have a larger database making extensive and intensive studies especially for rapidly mutating Y STRs covering different populations. The largest and widely used Y STR haplotype reference database is the YHRD.