**1. Introduction**

Around 5000 years ago, cultivated soybean (*Glycine max* [L.] Merr.) had been domesticated from wild soybean (*Glycine soja Sieb*. & *Zucc*.). This crop has originated in China, and it spreads gradually around the different parts of the world [1]. Soybeans is now one of the most economically important oilseed and biodiesel crops, as well as a major source of protein and oil for human and animal consumption [2]. Early soybean breeding relied primarily on farmers selecting preferred seeds from the planted population. Artificial hybridization has been used since the early 1900s. In the 1940s, North American breeding programs published the first modern soybean cultivar developed through hybridization [3, 4]. Artificial hybridization became more commonly used in soybean breeding, after that it is investigated that artificial hybridization dramatically enlarged the genetic basis of established lines and increased soybean adaption as well as productivity [5]. Soybean is largely crushed into soy oil

and meal, and it can be found in a variety of edible and nonedible goods, including cooking oil, animal grains, vegan food, and milk, as well as biodiesel and other industrial applications. Soybean oil is the most widely used cooking oil in the world, second only to palm oil [6].

The major objective of the most plant breeding projects/programs in soybean is to increase the yield and quality [2]. However, in the field of plant breeding, measuring primary traits such as yield or quality, which are mostly complex quantitative traits in a large breeding populations with thousands of genotypes, is time-consuming and labor-intensive [7, 8]. Due to genetic and environmental influences, breeding for yield is recognized to be a highly complicated and nonlinear process [9]. To this end, plant breeders can efficiently identify the promising lines at early growth stages using secondary traits for selection (e.g., yield component traits and reflectance bands), which are strongly correlated with the primary trait [10, 11].

The recent advances in sequencing technology have triggered a data boom in the biology field, propelling molecular biology into a stunning postgenomic *era*. From structural characterization to functional analysis, genomic research has progressed [12]. Despite the fact that genomic mapping, bioinformatics prediction, and other technologies aid in inferring gene function; however, any theory in life science requires ultimate confirmation. This inference is required for genetic transformation and vice versa; and it appears to be a powerful tool in functional genomics. Transgenic breeding is other important approach used to introduce genetic changes for specific plant traits. This method has been successfully used to increase crop productivity, production of biofuels, improve food quality and plant resistance against severe environmental conditions by breaking species limits. Furthermore, the implementation of genome-editing tools such as CRISPR/Cas9 relies on transformation procedures, demonstrating the necessity and importance of this technology.

Marker-assisted selection (MAS) has speeded up the breeding process especially in the production of disease and insect pest-resistant cultivars [13]. Linkage and physical maps are created using various types of genetic markers [14, 15]. Consensus Map 4.0 was created to combine known genetic and physical maps [16]. Large numbers of quantitative trait loci (QTLs) associated with different crop traits have been identified in soybean using genetic markers. However, the efficiency and precision of QTL location were restricted by limited number of molecular markers and their uneven distribution. To this end, the advances in the high-throughput genotyping and phenomics have greatly enhanced the precision and resolution in the gene mapping [17, 18]. Although, the advances in high-throughput genotyping were significant to alleviate the challenges in the plant breeding [7, 19, 20], but the advances in the high-throughput field phenotyping, is far lagging behind the genomics. Hence, the phenomics is a major bottleneck in current breeding programs [19].

The mechanism of genome editing technology is to introduce double-strand breaks (DSBs) within the genome at targeted sites using sequence specific artificial nucleases (SSNs), which are then repaired using nonhomologous end joining (NHEJ) or homologous recombination repair (HR) mechanisms, resulting in targeted mutagenesis by adding, removing, or replacing DNA bases [21]. Zin finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and clustered regularly interspaced short palindromic repeats/CRISPR-associated protein (CRISPR/Cas) are the most common SSNs at the moment [22, 23]. Despite their early development, ZFNs and TALENs are complex and expensive, which has limited their use. Since its inception, the CRISPR/Cas system has gained popularity in biological science due to its simplicity. The CRISPR/Cas9 system is the most well-known and has been

*Soybean Molecular Design Breeding DOI: http://dx.doi.org/10.5772/intechopen.105422*

increasingly used in the crop plants in the last few years [24]. The genome editing toolset has been broadened after the CRISPR/Cas9 system by selecting Cas9 orthologs and created variations [25–27]. Dead *Cas9* and *Cas9* nickases are two of them, and they have been employed extensively in base editing, gene expression regulation, epigenome editing, cell imaging, and other domains [28].

### **2. Marker-assisted breeding in soybean**

Marker-assisted breeding (MAB), also known as molecular-assisted breeding, is the use of molecular tools, primarily DNA markers, in conjunction with linkage maps and genomics to change and enhance plant as well as animal characteristics using genotypic tests [29]. The term MAB is used to explain the various novel strategies including MAS, marker-assisted backcrossing (MABC), marker-assisted recurrent selection (MARS), and genome-wide selection (GWS) or genomic selection (GS) [30]. MAB is recognized as a unique technique and a potent methodology for agricultural plant genetic modification, and it has been widely applied in a variety of crop species to date [29, 31].

Classical markers and DNA markers are two types of genetic markers used in plant breeding [32, 33]**.** Morphological, biochemical, and cytological indicators are examples of traditional markers. Random amplified polymorphic DNA (RAPD), amplified fragment length polymorphism (AFLP), microsatellites or simple sequence repeats (SSRs), restriction fragment length polymorphism (RFLP), and single-nucleotide polymorphism (SNP) are all examples of DNA markers (SNP). Marker-assisted breeding (MAB) is the most promising of the many applications of DNA markers in plant science for cultivar creation. MAB has huge potential to increase conventional plant breeding efficiency and precision by using DNA markers that are firmly related to critical genes or loci [33]. Several allele-specific functional markers for essential soybean features such as blooming and maturity, pod dehiscence, aroma, salt tolerance, soybean cyst nematode oleic acid content, raffinose content, and Kunitz trypsin inhibitor have recently been discovered [33, 34]. Phytic acid content, glycinin, conglycinin concentration, fragrance, and lipoxygenase were also discovered as strongly connected markers for seed nutritional value, which could help with the selection of novel varieties that are free of antinutritional chemicals [33].

MAB allows selection of plant features (that are expressed late in the plants genotype) at the seedling stage based on the genotypic data; hence, reducing the time it takes to identify the phenotypic of a single plant. MAS can swiftly remove unwanted genotypes for features that are displayed at later developmental stages. This trait is very significant and valuable for backcrossing as well as recurrent selection breeding programs, which require crossing with or between chosen individuals [17, 30]. MAB is unaffected by the environment, allowing selection to take place in any setting (e.g., greenhouses and off-season nurseries). This is particularly useful for improving qualities that are only expressed in the presence of favorable environmental circumstances, such as disease/pest resistance as well as stress tolerance [30]. MAS is based on reliable markers that are strongly connected to the QTLs associated with particular trait of interest and is more effective and efficient than phenotypic selection for low-heritability traits that are highly influenced by the environment (PS). In the heterozygous state, MAB utilizes the codominant markers (e.g., SSR and SNP) to allow effective selection of recessive alleles of desirable features. To detect quality

controlled by recessive alleles, no selfing or test-crossing is required; hence, MAB may save time and speed up breeding process [29].

The MAB method significantly accelerates the accurate and efficient introgression of targeted genes into recipient varieties, as well as the recovery of the recurrent parent genetic background. With just two backcrosses (BC2F2:3), markerassisted background selection in wheat was able to transfer *Yr15*, a stripe rust resistance gene in a recurrent variety and recover 97% of the genetic background of the recurrent parent, whereas phenotypic selection could only recover 82% of the background in BC4F7 plants [35, 36]. In this case, the MAB successfully saves the time it takes to obtain advanced breeding lines in half when compared with traditional approaches.

MAS and MABC have been frequently used to increase disease resistance and other relatively basic qualities [33, 37, 38]. MAB has been used successfully in a few soybean breeding programs to introduce single genic as well as polygenic traits into the desired genetic background (**Table 1)**. Moreover, MAB has been proven to be effective in improving quantitative features that contribute to soybean nutritional value, such as seed protein content and oil quality. MAB for seed protein content (SPC) in soybeans using SSR markers yielded up to 9% transgressive segregation in the trait after two cycles [48, 49].
