RAPDs are based on the PCR amplification of random DNA segments with primers of random nucleotide sequences that were inexpensive and easy to use. The primers bind to complementary DNA sequences and where two primers bind to the DNA sample in close enough for successful PCR reaction. The amplified of DNA products can then be visualized by gel electrophoresis [18] [19] .

RAPD markers have been widely used in diverse plant species for assessment of genetic variation in populations and species, fingerprinting and study of phylogenetic relationships among species and subspecies [20] . Nevertheless, disadvantages of RAPD markers are the fact that it predominantly provides dominant markers, and incapability to detect allelic differences in heterozygotes. Polymorphisms are detected only as the presence or absence of a band of a certain molecular weight, with no information on heterozygosity [21] . Additionally, because of their random nature of amplification and short primer length, they are not ideal for genome mapping. Moreover, these markers do not exhibit dependable amplification patterns and differ with the experimental conditions [16] .

During 1990s, Simple sequence repeats (SSRs) which is also known as microsatellites were established and provided a choice for many genetic researches since they are amenable to low, medium and high-throughput approaches. They are randomly tandem repeats of short nucleotide motifs (2 - 6 bp) [21] . SSRs are frequently highly polymorphic sequences normally present in animal and plant species [1] , and can be used to study the relationship between inherited traits within a species [22] . Microsatellite markers are often derived from non- coding/anonymous genomic regions, such as bacterial artificial chromosomes (BACs) and genomic survey sequences (GSSs). Therefore, development of SSR markers used to be expensive and laborious [3] . This assay is easily detectable by gel electrophoresis for few to hundreds of samples, which could be inexpensive by researchers with limited resources. Polymorphism is based on the variation in the number of repeats in different genotypes [23] . Since polymorphisms in longer penta-nucleotide and tetra repeats are easier to make a distinction in a variety of detection systems and longer repeats may be more robust [24] .

In recent years, SSR markers can easily be developed in silico due to the availability of large-scale gene (expressed sequence tag) EST sequence information for many plant species. Since EST sequencing projects have provided sequence data that is available in online databases and can be scanned for identification of SSRs [25] . The high degree of polymorphism as compared to RFLPs and RAPDs, their co-dominant nature and locus specific make them the markers of choice for a diversity of purposes including practical plant breeding. Therefore, (SSRs) have become a marker of choice for an array of applications in plants due to extensive genome coverage and hyper variable nature (Figure 2) [1] .

AFLPs are PCR-based markers, simply RFLPs visualized by selective PCR amplification of DNA restriction fragments. Such a marker is a multi-locus marker technique that combines the techniques of selective PCR amplification of restriction and fragments restriction digestion and it is possible to be applied into DNA of any origin [26] .

The technique involves three steps: Initially, oligonucleotide adapters are ligated to both ends of the resulting restriction fragments and genomic DNA is digested. Subsequently, the fragments are selectively amplified, using the adapter and restriction site sequences as primer binding sites for following PCR reactions. As the 3’ ends of the primers extend into the restriction fragments by (1 to 4 bp), only those fragments are amplified, whose ends are absolutely complementary to the 3’ ends of the selective primers. Therefore, only a certain amount of the restriction fragments is amplified. Finally, the amplified fragments are resolved by gel electrophoresis and visualized by either silver staining, autoradiography or fluorescence, resulting in a unique reproducible fingerprint for each individual [26] [27] .

The advantages of using this method are that it is cost-efficient, since a single assay allows detection of a large number of co-amplified restriction fragments and it requires moderate quantities of DNA. Additionally, higher levels of polymorphisms compared with RFLPs can be detected also AFLPs have much higher multiplex ratio (more markers per experiment) and better reproducibility than RAPDs. These features make this technology an attractive tool for saturating genomic region with low marker density and constricting genetic maps [16] [20] . However, AFLP assays have some limitations also. For instance, it requires the use of polyacrylamide gels for detection and it needs a greater technical skill. Moreover, most AFLP markers are dominant rather than co-dominant, because of the complex banding arrangements. In some cases the scoring of AFLP polymorphisms as co-dominant marker loci is possible, for the reason that diploid homozygous individuals cause more intense peak than heterozygous individuals [28] .

By group of researchers Colomba, Vischi [29] , genetic relatedness and identity of the durum wheat Graziella Ra, four Italian commercial durum cultivars (Grazia, Cappelli, Flaminio and Svevo and Kamut) were evaluated using (AFLPs). Their results revealed that the percentage of polymorphic loci within accession ranged from 6.57% to 19.71% (mean, 12.77%) and molecular variance was partitioned into 80% (variance among accessions) and 20% (within accession).

SNP is a single nucleotide base difference between two DNA sequences or individuals. SNPs are typically bi- allelic and arise either due to substitutions/point mutations (transversion and transition) or as a result of deletion/insertion of nucleotides and are detectable when similar genomic regions from different genotypes of different or same species are aligned [30] .

SNPs provide the simplest and ultimate form of molecular markers as a single nucleotide base is the smallest unit of inheritance, and therefore they can provide a great marker density. SNPs happen frequently in animals and plants. The probability to find polymorphisms in a target gene are increases due to high density of SNP markers which provides a huge advantage over previous markers that are at best closely linked to a locus of interest and not within [31] . In the case of linkage it can easily happen that a linkage is lost when a marker is applied to other populations with different recombination patterns. Typically, SNP frequencies are in a range of one SNP every (100 - 300) bp in plants. SNPs may present within coding sequences of genes, non-coding regions of genes or in the intergenic regions between genes at different frequencies in different chromosome regions [17] [32] .

Several methods are currently available for SNP discovery, either following the database approach, where SNPs are detected by following the experimental approach, or mining sequence databases, where genome regions of interest are screened for SNPs with one of various techniques established for the detection of SNPs. Moreover, it can be categorized into four reaction chemistries or principles: hybridization with allele-specific oligonucleotide probes, oligonucleotide ligation, enzymatic cleavage, and single nucleotide primer extension [32] - [34] .

In principle, the SNP methods show differences between a probe of known sequence and a target DNA containing the SNP site. The target DNA sections are typically PCR products and mismatches with the probe reveal SNPs within the amplified target DNA segment. The mismatching DNA segments can be sequenced then as the most direct way to identify SNP polymorphisms [35] .

SNP markers are likely to become the marker of choice for breeding in the near future, especially as the full sequences of more plant genomes will become available with the advantage of next generation sequencing tech- nologies (NGS) [31] .

More recently, it has become very-cost effective and easier to quickly identify a large number of SNPs in short time in any plant species. This was due to the emergence of the third generation sequencing. The advantage of this new sequence technology are expected to further reduce sequencing costs extremely to levels below $1 per mega base compared to $60, $2, and $1 expected costs for sequences generated by next generation sequencing [36] .

The large size of wheat genome has led to various approaches to reduce the cost of data production. These include the targeted re-sequencing of captured exome fragments and the establishment of confederations to share the cost of genome sequence data generation [37] [38] . In a different study, by using illumina sequencing of cDNA libraries, 14,078 putative SNPs were recognized across representative samples of UK wheat germplasm, with a proportion of these SNPs validated using KASPar assays [39] . In addition, in developed countries such as UK, France and Australia many attempts for large-scale SNP development in bread wheat were undertaken, leading to the development of millions of SNPs. These SNPs will be widely used for molecular breeding in wheat [40] .

KASPar genotyping may be of specific interest to researchers and breeders who are interested in analysing a small number of targeted SNPs in a large number of samples. This makes KASPar a cost-effective, simpleand flexible genotyping system; since the assays can be modified with a range of DNA samples and it does notrequire a hybridization step; as an alternative it includes real-time detection of the product [41] .

The chemistry of KASPar assyas involves one common reverse primer and two competitive allele specific tailed forward primers. To determine the alleles at a specific locus, this assay system relies on the discrimination power of a novel pattern of competitive allele specific PCR [3] . In such cases, this assay involves competitive allele-specific PCR for a given SNP, followed by SNP detection through Fluorescence Resonance Energy Trans- fer (FRET) [41] .

This assay for the target SNPs has been developed and used for genotyping commercially by Kbioscience UK (http://www.kbioscience.co.uk/). This company perfected this technique to improve the performance of the detection platform by incorporating a 5’ - 3’ exonuclease cleaved Taq DNA polymerase and a homogeneous Fluorescence Resonance Energy Transfer (FRET) detection system. The two allele-specific primers of a SNP are designed so that they incorporate with a unique (18) bp tail to the respective allele specific products, which in later cycles allow incorporation of allele specific fluorescent labels to the PCR products (with the help of corresponding labelled primers) (http://www.kbioscience.co.uk/). The mechanism of KASPar chemistry has been presented in (Figure 3) [42] .

In KASPar assays there is no need of sequencing to identify SNPs, instead SNP flanking sequences already known while developing different types of genotyping assays (for instance, illumina) can easily be used for primer design (one common and two allele-specific primers) for KASPar assays [34] .

Despite the fact that KASPar genotyping assays have come to the market very recently, they have started to be used for a large number of commercial species. In maize, a set of 695 highly polymorphic gene-based SNPs from a total of 13,882 GG-validated SNPs were selected and converted into KASPar genotyping assay with a success rate of 98% [43] . Additionally, in wheat, the technique has been used for constructing a linkage map containing several hundred SNPs [39] .

With the increased throughput of NGS platforms, re-sequencing for genome-wide surveys of genetic diversity became reasonable [44] . However, this assay is bioinformatically challenging, impartial estimation of genetic diversity across the genome in both coding and non-coding regions could be determined. Additionally, it allows for the detection of various types of genetic variation; this detection capability contains not only SNPs and small indels, but also large mega-base scale indels [45] .

This assay involves the use of restriction enzymes for reducing the complexity of genomes followed by targeted sequencing of reduced proportions, in that way each marker can be sequenced at high coverage across many individuals at low cost and high accuracy [3] . A workflow of GBS has been presented in (Figure 4).

The main target for constructing GBS libraries was based on reducing genome complexity with restriction enzymes, which may reach important regions of the genome that are unreachable to sequence capture approaches [46] . The procedure has been demonstrated with barley (Oregon Wolfe Barley) at the recombinant inbred lines populations and maize (IBM) where about (25,000 to 200,000) sequence tags were mapped, respectively. With this method, species that lack a complete genome sequence can have a reference map settled around the restriction sites, which can be done in the process of sample genotyping. This system has been adjusted for reducing missing data points and improved SNP calls [2] .

Gene pyramiding is defined as an assembly of multiple desirable genes which can be combined into a single genotype from multiple parents. This is referred as one of the major applications of marker assistant selection, as gene pyramiding via conventional plant breeding is difficult, if not impossible [47] .

The methods for pyramiding favourable alleles can be used in the same way to accumulate QTL controlling different traits. A main difference in the model is that alleles at different trait loci to be accumulated may have different favourable directions, for instance negative alleles are preferable for some traits but positive alleles are favourable for others. As a result, to meet breeding objectives one may need to combine the positive QTL alleles of some traits with the negative alleles of others [4] .

Selection for multiple traits may be completed in one cycle if the population size is large enough to allow desirable individuals to combine different traits (see Figure 5). Nevertheless, the number of trait loci that can be manipulated in one cycle is restricted because the population size required covering the recombinants increases exponentially with the increase of the number of traits/loci [4] .

To overcome this limitation, a strategy was proposed by Bonnett, Rebetzke [48] . In this method, individuals has been selected by all target markers for both heterozygous and homozygous forms to gain a subset of population that contain higher frequencies of the target alleles so that to obtain the homozygotes at the target loci, a smaller population size is required in the following generation.

For Barley Yellow Mosaic Virus complex, a variety of markers have been developed for selection of the rym5 and rym4 resistance genes on chromosome 3H [49] . In another study, the regions of a typical Spanish barley line that have VRNH2 and VRNHl were introgressed into a winter variety. A set of 12 lines introgressed with all four possible combinations of VRNHl and VRNH2 has been assessed for frost tolerance and vernalization requirement [50] . In practical, what has to be taken into account when applying such strategies pyramiding has to be repeated after each crossing, since the pyramided resistance genes are segregating in the progeny [51] .

In the 1990s, marker-assisted recurrent selection was proposed, which uses markers at each generation to target all traits of importance and for which genetic information can be achieved. Parents contribute different favourable alleles when the QTL mapping is conducted based on a bi-parental population. Therefore, the perfect genotype is a mosaic of chromosomal segments produced by recombination between the two parents [47] .

MARS refers to the improvement of an (F2) population by one cycle of marker-assisted selection (for instance, based on data marker scores and phenotypic) followed usually by two or three cycles of marker-based

selection (for instance, based on marker scores only) [4] . It is possible today to define an ideal genotype as a pattern of QTLs, all QTLs carrying favourable alleles from various parents (Figure 5). After several successive generations of crossings, it might be possible to get close to the ideal genotype. In addition, this system can start without any QTL information, and selection can be based on significant marker-trait association established during the MARS process [6] .

Studies have revealed that, in accumulating favourable alleles, (MARS) was superior to phenotypic selection [52] . Moreover, through MARS in maize, the rates of genetic gain accomplished for complex traits were about twice those of phenotypic selection in some reference populations [53] . The usefulness of including prior knowl- edge of QTL under genetic models has been studied that included QTL number, gene effects, heritability, epistasis and linkage. It is concluded that with known QTL, MARS is most favourable for traits controlled by a large number of QTL [54] .