Crop germplasm: Molecular and physiological perspective towards achieving global crop sustainability

Germplasm is a long-term resource management mission and investment for civilization. For both food and nutritional health, the present changing environmental scenario has become an urgent universal concern. Multiple excellent studies have been previously performed, although the advancement and innovation of practices will require the exploration of the potentiality of crop germplasm. In this study, we emphasized (i) germplasm activates, current challenges and ongoing trends of the crop germplasm, and (ii) how the system biology will be helpful to understand the complex traits such as water use efficiency (WUE), and nitrogen use efficiency (NUE) to mitigate challenges for sustainable development under growing food requirement and climate change conditions. We focused on a vision for transforming PGR into a bio-digital resource system, for the development of climate-smart crops for sustainable food production. Moreover, this review attempted to address current challenges, research gaps and describe the advanced integrated strategies that could provide a platform for future crop improvement research. Keyword: Artificial intelligence, Core set, Climate change, Nitrogen use efficiency, Omic approaches, Plant genetic resources, Stress, Systems biology, Water use efficiency

circumstance, for future life insurance, a vision has been proposed for transforming PGRs into a bio-digital resource system. To harness the benefit of PGRs, despite well-equipped maintenance and storage facilities, the additional execution of frontier technology transfer is lacking.

Germplasm activity, trends and current challenges
The breeding resource maintenance program is the first important component of digging into the PGRs. Figure 1 shows the structural component and activity of germplasm are divided into four major components. To harvest the benefit of PGRs, the development and maintenance of a precise "core-set" collection has been illustrated. The core-set collections should have the greatest diversity of total genetic resources (comprising both exotic and endogenous collections) because they served as a reference population for additional breeding program (Mascher et al., 2019). The core-set collection should be free of genetic duplicates and must have evolutionary potential genotypes. Basic cytogenetic information such as ploidy level, mitotic behaviour, and genetic diversity/redundancy is required for an additional breakthrough. Such a core-set will be useful to understand the genetic variation in the species that can be used for innovative field experiments and subsequent parental material in the breeding program. Moreover, the cost of maintenance of germplasm could be also be reduced. mutation population mapping are preferably powerful strategies for establishing genetic trait association. Phenome QTL (phQTL), metabolome QTL (mQTL), proteome QTL (pQTL) and expression QTL (eQTL) have been extensively used to discover genes and functional annotation. (D) Integrated Breeding Approaches: Forward breeding, MAS/MABC, PAGE, and transgenic approach can be employed to integrate the simply inherited trains in which genomic selection can be employed to integrate the complex inherited trains. The aim of global sustainability will be fulfilled by this systematic evaluation of genetic resources that will simultaneously save time and budget expenditure. Abbreviation: WUE, water use efficiency; NUE, nitrogen use efficiency; DTI, drought stress index=(T/C)*100; C, control; In the context to explore genetic resources, (1) sink of genetic diversity with increase in the number of gene bank accessions; (2) accumulation of duplicate accessions over a longer period collection; (3) degradation of collection, particularly exotic accessions in response to the edaphic/biotic-abiotic factors; (4) loss of originality because of long-term domestication; (5) insufficient knowledge for characterizing complex important treats; (6) accumulation of rare deleterious alleles during crop domestication; and (7) low recombination frequency between the wild relatives have been considered the major bottleneck for crop improvement (Langridge et al., 2019). Hence, clearly, new and innovative utilization strategies are required to reduce the complexity of these limitations. Therefore, for crop improvement, it is important to intelligently guide the germplasm resources. Accordingly, we should focus on a vision for transforming PGRs into a bio-digital resources system. Currently, systems biology is considered as a potential approach for the complete understanding of biological systems (Lavarenne et al., 2018). Integrated phenomics and nextgeneration sequencing (NGS) technologies have been utilized for identifying genetic markers associated with desirable traits in multiple crops, including rice, foxtail millet, pigeon pea, pearl millet, cotton, rapeseed, chickpea, and grape. Moreover, the large-scale re-sequencing of germplasm accessions have been re-sequenced in rice (3010), pearl millet (994), and chickpea (429) with the ongoing and deep profound of sequencing costs (Varshney et al., 2020). This study suggests that large-scale sequencing should be undertaken for all available accessions. Such projects will generate 'big data' that has storage and computational challenges. Therefore, the advancement of sequence-based study in next-generation breeding, high-throughput bioinformatics platforms, and passionate scientists is required. These datasets will provide the information to breeders for mining superior alleles/haplotypes that can serve as the key to select precious parental material for breeding.
Breeders mostly target functional traits, such as water use efficiency (WUE) and nitrogen use efficiency (NUE), to improve climate resilience crop productivity (Tracy et al, 2020).
Agriculture is the world's largest consumer of water and accounts for 70% of all water used worldwide; by 2050, global food demand is expected to increase by 70% (Connor et al. 2017). The ongoing global catastrophe of water is one of the major problems in the presentday climate scenarios and drought is considered as the most severe abiotic stress affecting crop productivity at a global level. Moreover, nitrogen is the most essential nutrient for plant's growth, development, and yield. Increasing trends of fertilizer application to achieving productivity, resulting in the decreasing the acquisition and subsequent utilization of applied nitrogen in crop plants. Additionally, the surplus amounts of inorganic and organic nitrogen fertilizers often imposed drastic negative impacts on the environment (Kumar et al., 2020). For example, NUE for cereal production is ~33% globally and the unaccountable 67% corresponds to a $15.9 billion loss/year of nitrogen fertilizer (Raun et al., 1999;Fagodiya et al., 2020). Moreover, it is estimated that ~1% increase in crop NUE could annually save $1.1 billion (Kant et al., 2011;. Hence, for achieving greater agricultural sustainability, the development of crop plants with drought-tolerant and more efficient nitrogen usage is, therefore, an important research challenge. Unfortunately, to date, extremely few studies has been conducted ( Table 1). necessary. Therefore, to quantify the potentiality of genotypes, next-generation trait-based evaluation should be performed on field conditions. This will help us identify superior genotypes/accessions that are stable, abundant, and environmentally less sensitive (Martin et al., 2015). For identifying the desired genes and associated treats, the recent trend of the functional phenomic approach will be helpful (Braun et al., 2020). The availability of these resources will be accelerated by increasing systems biology approaches to understand the molecular mechanism of complex traits such as WUE and/or NUE (Figure 2).

Figure 2:
A comprehensive gene discovery strategy of water use efficiency (WUE) and nitrogen use efficiency (NUE) associated traits. Synthesized high-throughput multi-omic integration data will confirm the phenotypic behavior, mutational landmark (SNP), expression pattern, the abundance of protein, and metabolic signature with associated treatment. Different transcript, metabolic, protein-protein interaction libraries of different genotypes will be helpful for enrichment and functional network analysis. Previously reported different data-mining bioinformatics-based analysis, e.g., (a) NGS-based RNAseq study will be helpful to identify abundant genes, (b) multidimensional clustering and gene network analysis of proteomic study (2D gel electrophoresis and co-immunoprecipitation, etc.), and (c) functional phenomics that will help to predict and discrimination of the desired gene with the associated trait. The identification of a gene that is strongly associated with a particular phenotype, e.g., GENE 1 strongly associated with those plants that have high NUE in low N 2 and dehydration conditions can be analyzed. Finally, big-data enrichment and functional network analysis by artificial intelligence (AI) will provide the opportunity to develop different types of models such as random models and guided models. The guided model will be more precious and accurate, which may develop as per our requirement, whereas a random model represents the overall analysis. Thus, an intelligently guided model will help in the understanding of complex traits with genetic background, G*E interactions patterns, developing ideotype breeding especially for underground traits, identify superior parental lines for promising breeding, improving smart farming.

System biology and future sustainability
To understand the genetic basis of trait variation, NGS-based genome-wide association studies (GWAS) are a powerful tool. This approach has been successfully applied to many important traits in different species, including yield-relevant traits, in crops. Sufficiently powered GWAS often identify tens to hundreds of loci containing hundreds of singlenucleotide polymorphisms (SNPs) associated with a trait of interest (McMullen et al., 2009).
Once traits are correlated with specific biosynthetic/metabolic pathways and desired alleles have been identified, researchers can take up a broader understanding of crop biology to anticipate parental and allelic combinations that will discover superior agronomic traits.
Although simply analyzing the genome sequences or genome-wide association study will become a "null hypothesis", we will never get the account of gene regulation from gene/protein sequences. To date, the identification of candidate genes detected by transcriptomics approaches and/or mapping is not available in most crops (Li et al., 2019).
Furthermore, it is important to understand in-depth molecular mechanisms of their possible agronomic values. In addition to structural genomics, information on gene expression atlases, epigenome maps, proteome maps, and metabolome maps have been developed for a few crop species (Zogli et al., 2020). From the last ten years, plant molecular geneticists efficiently practice the most sophisticated toolkits for digging into valuable genetic resources that have been used such as advanced cloning strategies, genetic transformation methods, different gene-editing tools, co-immunoprecipitation, two-hybrid system, mass spectrometry, and datamining bioinformatics. In this context, the recent study published by Jia et al. (2020) (Watson et al., 2018).
Understanding the functional aspects, the conventional genetic methods-integrated strategy has the following advantages: (a) less tedious because of the multiplication of population is not required; (b) large plant sample size, mutants, and stable homozygous overexpressed plants are not required; (c) low-cost of mining of publicly available database resources makes the comparative study of omics data easier; (d) an integrated approach could identify and provide comprehensive interactions, co-expression, and metabolic/biosynthetic grid; and finally (e) integration of advanced artificial intelligence (AI) gives the sufficient multidimensional data. Thus, this integrated approach should turn out to be one of the most efficient systems for a high-resolution understanding of the complex genetic mechanisms in the near future. For crop improvement, the recent trends of genome editing toolkits have significant potentiality. As a plant geneticist/breeder, our aim is to focus on the exploration and strengthen the advanced strategies to harness the benefit of ~7.4 million gene bank germplasm material. Our present brief discussion will provide the advancement of germplasm assets that will meet the required sustainability in the near future.