**6. Conclusion and future prospective**

Traditional breeding has made important contributions to soybean improvement and the generation of soybean varieties with enhanced yield, quality, and tolerance to numerous stressors throughout the last century. Traditional crop development technologies, on the other hand, are unable to keep up with the world's rapidly rising population and climate change [166, 167]. Reduced generation time allows for faster soybean breeding, which may be accomplished by the quick creation of homozygous lines employing the doubled-haploid (DH) production process. The creation of a high-throughput DH production program in soybean would be tremendously beneficial in achieving the crop's targeted genetic gain. Soybean androgenesis, root development, and unusual shoot induction have all shown slight advancements. However, there is currently no efficient, repeatable way for producing doubled haploids in soybean. One of the primary impediments to the development of a commercial DH production procedure in soybeans may be the tissue's resistance to *in vitro* regeneration [168, 169]. To achieve a sustainable yield, it is necessary to identify genetic resources in the form of water-stress-resistant soybean genotypes and genomics-assisted waterstress mitigation approaches. Several techniques, such as QTL mapping, genome-wide association mapping, and comparative transcriptome analyses, are being used to determine the genetic basis for water-stress tolerance in soybean [170, 171].

Recent advances in NGS-based genotyping technologies and powerful computational pipelines have significantly reduced the cost of WGS/WGRS, allowing the

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

discovery, sequencing, and genotyping of hundreds of thousands of markers in one step. For large-scale marker identification, NGS-marker technologies based on reduced representation sequencing are the ideal solution, especially for the huge and complex soybean genome. These NGS-based marker approaches represent the soybean's partial genome, and they can even be used without a reference sequence. RAD-seq (or its variations) and GBS are two NGS technologies that have previously proven to be efficient and effective procedures for GAB and have been widely employed for GS investigations in various agricultural plants. Furthermore, the NGS has enabled the fabrication of high-density SNP chips for HTG in soybean. The low cost, genome-wide marker coverage, better speed and throughput, and higher marker density of NGS-marker technologies have allowed geneticists to explore the inheritance of numerous traits at the nucleotide level accuracy.

On the other hand, GS employs a number of markers spread over the entire genome to forecast the breeding value of a breeding line for selection. GS can quantify Mendelian sampling without phenotyping the entire population thanks to genomewide dense markers. It shortens cycles to save time while also increasing genetic gain per unit of time. GS was compared with traditional phenotypic selection in soybean to see if it has any advantages in terms of accuracy and time savings.

GE technologies, particularly CRISPR-based systems, have advanced quickly, with the majority of them being implemented to give effective tools for soybean improvement. If this technique is properly implemented in plant breeding programs, a recent field trial of high oleic soybean employing TALENs has indicated the bright future of soybean improvement. Currently, the discovery of more GE target genes associated with agronomically important traits, the adoption of newly developed GE technologies, the simplification and renovation of editing reagent delivery, and the improvement of target mutant recovery method in soybean will improve editing outcomes, save time, and lower product development costs. The development of GE products will be aided by the cost-effective preparation of intellectual property for GE technologies, as well as breeders' and farmers' comprehension of GE-related government regulation. In several countries, transgene-free or DNA-free edited plants are considered nongenetically modified events, making GE soybean production easier. In future, more applications of "base editing" for single genes or several genes at once would substantially aid functional research and molecular design breeding in soybean.

Next-generation GAB in agricultural plants has been enabled by recent advances in crop phenomics and genomics, which have provided several high-throughput platforms, as well as statistical approaches and computational tools. When these modern technologies are integrated, they can precisely and accurately identify genes/ QTLs, as well as their beneficial usage in soybean breeding [172, 173]. Despite the fact that high-throughput SNP genotyping technologies have completely revolutionized marker application in soybean breeding, they have enabled research groups to apply GWAS and GS for soybean improvement on a regular basis. These marker technologies, however, must be paired with HTP to produce meaningful genetic gain from complex features in order to reap the full benefits of genomic investigations. So far, only a few studies involving the use of both HTP and HTG in soybean have been reported. This is because large-scale field-based HTP has a greater cost. New advances in crop phenotyping technology are expected to make HTP more inexpensive for commercial application in soybean breeding projects in the near future. This would undoubtedly increase the scope of germplasm assessment and facilitate the development of better soybean cultivars. WGRS-based genotyping will become increasingly

viable and cost-effective as the cost of DNA sequencing falls. Sequencing-based genotyping employing genome-reduction methods such as GBS and RAD-sequencing appears to be more cost-efficient for breeding-based applications such as GS at the moment. Since the cultivated soybean has a limited genetic basis, genome editing and TILLING can be used to produce a variety of changes in these orthologs, from knockdown to knockout alleles. For quick deployment of these alleles in breeding programs, it should be combined with the speed breeding facility.

The ZFNs, TALENs, CRISPR/Cas9, CRISPR/Cas12a, BEs, and other CRISPR/Cas variations provide a robust genome editing toolkit that will aid future functional genomic and genetic improvement studies in soybeans and other plants. CRISPR/ Cas9 technology may become the preferred method for soybean breeding due to its efficiency, multiplex editing, and high-throughput mutagenesis capabilities, as well as its maturity. With the progress of additional genome editing methods, however, soybean genome editing will become more versatile. Despite the fact that substantial effort may be required to employ these techniques, given the enormous potential of genome editing and the economic importance of soybean, we anticipate that these issues will be resolved in the near future.
