1 Introduction

In many developing countries, common beans (Phaseolus vulgaris L.) are not only a common food on the dining table but also a source of nutrition for many families. It is rich in protein, carbohydrates, minerals and vitamins, which can not only fill your stomach but also provide nutrition. It is precisely because of their strong adaptability and high nutrition that common beans are becoming increasingly crucial in global food security and the fight against malnutrition. The phytochemicals and antioxidant components it is rich in are also believed to help enhance immunity and reduce the risk of chronic diseases.

Of course, relying solely on natural characteristics is not enough. Under the background of sustainable agricultural development, people have begun to pay more attention to how to improve the seed quality of common beans. This is not a new topic. Early breeders focused more on yield and disease and pest resistance. As for the "soft indicators" such as flavor, nutrition and stress resistance, they are often overlooked instead. The problem lies in that these complex traits are influenced by both genetics and the environment, and it is difficult to effectively screen and improve them through traditional breeding methods. Moreover, it is no easy task to accurately capture the metabolic phenotypes of these traits.

Fortunately, technological means are gradually opening up breakthroughs. With the development of high-throughput transcriptomics and metabolomics, we can now track gene expression and metabolic accumulation more comprehensively. If these "omics data" are integrated and analyzed, it is possible to clarify the genetic and biochemical mechanisms behind seed quality traits, identify key genes or metabolic pathways, and in turn guide more precise variety improvement. For instance, in common beans, metabolic pathways such as phenylpropanes and flavonoids have been found to be closely related to their nutrition and stress resistance. These findings provide new theoretical support for breeding strategies (Scossa et al., 2020; Zhang et al., 2023).

We plan to review the existing research on integrating transcriptome and metabolome data to improve the quality of common bean seeds through this study. Starting from the significance of traits, then moving on to the difficulties of traditional breeding, followed by the advantages and research progress of multi-omics approaches, and finally looking forward to the potential of these technologies in enhancing nutrition and improving agronomy.

2 Seed Quality Traits and Their Determinants in Common Bean

2.1 Nutritional traits

In many regions that rely heavily on plant protein, common beans are one of the important sources on people's dining tables, especially when animal protein is hard to come by. Its seeds are rich in protein, with a content of approximately 17% to 30%. However, the nutritional differences among different varieties are indeed quite significant. Some genotypes have a high protein content and are often accompanied by a relatively large amount of iron and zinc-two trace elements that are very crucial for human health. Interestingly, some varieties of common beans perform quite outstandingly in this regard. In addition, the starch in common beans should not be overlooked. It not only comes in a wide variety but also directly affects the calorie content. In addition to these, common beans also provide many trace elements, such as calcium, magnesium, potassium, phosphorus, etc. The content of these elements varies greatly in different germplasm resources. It is precisely this "difference" that gives us the opportunity to pick out those varieties with better nutritional performance, laying the foundation for subsequent breeding (Lei et al., 2020; Jan et al., 2021).

2.2 Anti-nutritional and functional metabolites

By the way, being nutritious doesn't mean there are no problems. There are also some less "friendly" components in common beans, such as tannin, phytic acid, polyphenols, etc., all of which belong to anti-nutritional factors. They can combine with minerals or proteins, thereby affecting human absorption. In some varieties of white beans, the content of starch and uronic acid is particularly high. Such components may instead cause discomfort in digestion. However, the situation is not so absolute. For instance, polyphenols actually have antioxidant effects and can be beneficial to health to some extent. So, the issue is not whether these substances exist or not, but how to strike a balance between "reducing negative impacts" and "preserving benefits". This is precisely a rather challenging point in the breeding of common beans (Aziziaram et al., 2022).

2.3 Processing-related traits

Whether you are a researcher or a consumer, when it comes to common beans, you probably can't avoid one question: Are they easy to cook? The cooking time, texture and flavor directly affect whether people are willing to eat it or not. The duration of cooking is related to traits such as the thickness of the seed coat, water absorption capacity, and seed hardness, most of which are genetically controlled. Now, QTLS (quantitative trait gene loci) related to these characteristics have been identified, and they are usually hereditary, meaning there is an opportunity to improve them through selection and breeding. However, when it comes to texture and flavor, the matter is not so simple. Genetic factors play a part, and environmental conditions are also crucial. Some varieties perform particularly well in this aspect, such as cooking quickly and having a good taste when eaten. It is worth noting that these processing traits are often associated with nutrition and seed appearance (Figure 1), which also reminds us that when making variety improvements, we should consider multiple aspects rather than just focusing on a single trait (Sandhu et al., 2018; Dias et al., 2020; Nadeem et al., 2020).

Figure 1 Variation in seed testa color in common bean germplasm. The figure shows the diverse nature of genotypes. A=French yellow; B=Canadian red; C=cream; D=olivaceous; E=light yellow; F=black eye; G=black; H=brown; I=light gray; J=light brown; K=great northern; L=red (Adopted from Jan et al., 2021)

3 Transcriptomic Insights into Seed Development and Quality

3.1 Gene expression dynamics during seed maturation

The process of seed development is not static. In fact, it involves the dynamic expression of a large number of genes, especially in different tissues. For crops like legumes and cereals, it has been discovered through time transcriptome analysis that thousands of types of genes are involved from early cell division to late accumulation of stored substances (Yi et al., 2019; Yu et al., 2023). Of course, not all genes are active at the same time or in the same tissue. For instance, there are obvious differences among the embryo, endosperm and seed coat, each activating different gene modules. In the early stage, it is mainly related to cell division and differentiation, while in the later stage, the expression of genes related to protein storage, starch synthesis and coping with stress begins to increase (Chen et al., 2014). The changes in these stages actually laid the foundation for the subsequent seed quality.

3.2 Regulatory networks linked to nutrient accumulation

Regarding the accumulation of nutrients, it is not a simple piling up but rather regulated. Through co-expression analysis, researchers identified some regulatory modules related to proteins, starches, lipids, etc. (Rangan et al., 2017; Zhang et al., 2021). However, the expression of position and time is not arbitrary. Genes that control carbon and nitrogen metabolism, or pathways related to the synthesis of flavonoids, saponins, and phytic acid, are often preferentially activated during seed development (Song et al., 2022). These networks do not operate independently either. They coordinate with each other to ensure that nutrients are rationally distributed during the critical stage of seed filling, meeting the needs of plant development while also influencing the quality in the later stage.

3.3 Key transcription factors modulating quality-related pathways

When it comes to seed quality, one cannot fail to mention the core transcription factors that regulate it. Several families are relatively common, such as bZIP, MYB, MADS-box, NAC, WRKY and bHLH (Manzoor et al., 2025). However, the directions in which they act are different. Some, like bZIP, are more inclined to regulate the accumulation of proteins and lipids, while MYB and MADS-box are more related to organ formation. WRKY is even more special, involving both development and defense responses (Deng et al., 2018). More importantly, their expression not only depends on space but also on chronological order, which is equivalent to controlling which downstream genes should "come online" at what time, thereby influencing the final quality of the entire seed.

4 Metabolomic Profiling of Seed Composition

4.1 Primary metabolites shaping energy and nutritional reserves

The accumulation of nutrients in seeds is not a single or continuous process. Primary metabolites such as amino acids, sugars, organic acids and fatty acids appear at different rhythms during seed development. Some substances, such as proteins and oils, mainly accumulate in the later stage of grouting. But starch reaches its peak very early and then decreases. Similar patterns have also been discovered in studies of leguminous plants and other crops (Li et al., 2015; Ding et al., 2023). Metabolic processes such as glycolysis, the TCA cycle, and amino acid synthesis may seem complex, but their ultimate goal is clear-to transport carbon and nitrogen to the seeds, providing energy and nutrients for the growth of seedlings and human diet in the future.

4.2 Secondary metabolites influencing flavor, anti-nutritional factors, and health benefits

Not all components that are useful to the human body belong to the "main players". Secondary metabolites, such as flavonoids, saponins, alkaloids, tannins and phenols, do not directly participate in growth, but are crucial in terms of taste, color and health benefits. The types and contents of such substances are not fixed and are greatly influenced by genotype and the stage of seed development. For instance, dark-colored seeds usually have stronger antioxidant activity because they often contain more flavonoids and certain essential amino acids. However, problems are not absent. Some substances, such as phytic acid and tannin, can hinder the human body's absorption of minerals (Farag et al., 2019; Li et al., 2019; Solanki and Abhyankar, 2024). This makes breeding a bit of a dilemma-enhancing nutrition and reducing anti-nutritional factors may not be achievable simultaneously.

4.3 Environmental modulation of metabolite diversity in bean seeds

Even if the seeds come from the same species and are planted in different places, the results may be completely different. Environmental conditions, such as genetic background, seed coat color, and even climate, can all affect the types and quantities of metabolites. Some subgroups will accumulate more functional metabolites when facing drought or poor soil quality. However, under stable environmental conditions, they may perform averagely (Campbell et al., 2021; Dossou et al., 2021; Kim et al., 2024). This indicates that metabolome data cannot be judged merely on the surface; it is more necessary to understand it in combination with the planting environment. It is precisely for this reason that applying such data to breeding programs makes it possible to screen out varieties that can perform stably and have ideal quality in real environments.

5 Integrating Transcriptome and Metabolome Data

5.1 Correlating gene expression patterns with metabolite accumulation

Researchers often attempt to look at gene expression and metabolite changes together. Especially under the same conditions, if some genes are always expressed together or the patterns of certain metabolites always change accordingly, there may be some kind of connection between them. Through these clues, a network of relationships between genes and metabolites can be drawn to understand the possible regulatory logic behind them. Especially for those metabolites that affect seed quality, people are more eager to find the expression genes closely related to them (Yin et al., 2022; Sanches et al., 2024). Of course, such a match cannot be found casually. Usually, it requires an integrated strategy to achieve.

5.2 Network and pathway-level integration approaches

When transcriptome and metabolome data are viewed together, they can also be analyzed from the perspective of pathways and networks. It is not just about looking at which genes or metabolites have changed, but rather figuring out which biological pathways are "activated" or "silenced". This is what pathway enrichment analysis does. It can indicate which metabolic pathways are significantly active in both omics data. Another type of method, such as genome-scale metabolic modeling, takes into account the changes in metabolic fluxes and simulates how many metabolites are produced under different expression conditions. In addition, there are some network methods such as similarity fusion or interaction group analysis, whose goal is to unify the associations in different data into a single network graph and identify complex but important intersection points (Wanichthanarak et al., 2015; Cavill et al., 2016; Di Filippo et al., 2021; Jendoubi, 2021).

5.3 Discovery of candidate biomarkers for seed quality traits

Often, the ultimate goal of research is to find something that can be applied to breeding. Those metabolites and genes that can predict seed quality are the biomarkers that breeders desire. Whether it is a linear model or machine learning, as long as it can be seen that a certain trait has a strong correlation with the combination of a certain group of genes and metabolites, it is worth noting. Path and network analysis can further help us identify key nodes, which might be the regulatory switches. Ultimately, these markers can be used in selective breeding or as targets for genetic engineering-as long as the goal is to improve the quality of seeds (Siddiqui et al., 2018; Patt et al., 2019).

6 Case Study

6.1 Comparative synthesis

In the study of legumes, black beans are actually a category that has been studied in greater detail. Traits such as yield and canning quality have been measured by many researchers using breeding materials from different sources, and various global germplasm banks have also been utilized incidentally. One black bean study is quite typical. It combines genomic information with near-infrared spectroscopy (NIRS) to see if it can predict which one has a higher yield and better canning traits. The effect is indeed much more accurate than previous models, especially after using the new genotype, the accuracy rate has significantly improved (Izquierdo et al., 2025). However, it's not just black beans; there are also many similar practices for broad beans and soybeans-quality characteristics such as protein, oil, and anti-nutritional components can all be evaluated through multi-omics methods. High-throughput phenotypic analysis combined with omics integration can provide a clearer view (Figure 2).

Figure 2  Comparative maps of ortho-MQTLs among common bean, soybean and pea. MQTL, meta-analysis of quantitative trait locus; MQTL-YC, meta-analysis of QTL for seed yield components (Adopted from Izquierdo et al., 2023)

6.2 Convergent findings

Some paths have repeatedly appeared in many studies, regardless of the type of bean. Processes such as cell wall-related remodeling, amino acid synthesis, or the metabolism of phenylpropane-like substances are always closely related to seed quality. Not all traits can be directly attributed to these pathways, but common indicators such as texture, protein levels, and antioxidant capacity are always associated with them. Moreover, once the data of the transcriptome and metabolome are integrated, it will be found that these key genes and metabolites are indeed related to the composition of seeds, and many times they are the breakthrough points for marker development and breeding material selection (Yang et al., 2025).

6.3 Methodological lessons

To make multi-omics research more reliable, the details of the methods really cannot be ignored. For instance, in the case of sampling, if data from different periods are not covered, many dynamic relationships will be missed (Chaudhary et al., 2015). In addition, batch effects and standardization processing are also crucial. Once these steps are relaxed, omics data will become difficult to compare (Lippolis et al., 2024). Another point worth mentioning is the model. Nowadays, multi-feature fusion and network-based models, combined with machine learning methods, can indeed enhance prediction accuracy, especially when multiple genotypes are updated in the training data, the effect is more obvious.

7 Challenges and Opportunities

7.1 Technical Hurdles

Multi-omics sounds advanced, but when it comes to specific operations, problems arise. The design schemes used in different experiments are not uniform, the data processing procedures are diverse, and even the versions of the reference genomes are inconsistent. All these make cross-comparisons in the study of common beans very troublesome. The accuracy of the identification of candidate genes and metabolites is thus greatly reduced. Another challenge in metabolomics is that the annotation of metabolites is still incomplete, and many key compounds simply cannot find corresponding items in the standard library. To put it bluntly, there are still many gaps in our understanding of the metabolic composition of seeds. In addition, the lack of a unified standard for phenotypic data, coupled with the interference of batch effects, makes it difficult to integrate multi-omics data, and the interpretation results are also prone to bias (Izquierdo et al., 2023).

7.2 Biological complexity

The quality of common bean seeds is not determined by just one or two genes; the underlying regulatory mechanisms are very complex. The interaction between genes and the environment is the decisive factor. External conditions such as climate, soil and planting methods can change the gene expression pattern with the slightest movement, and even the metabolite profile will follow suit. Many studies have thus fallen into the awkward situation of "accurate today but not tomorrow". To solve this problem, it is necessary to rely on multiple environmental experiments and evolutionary analyses to understand the true range of trait changes as clearly as possible and identify those genotypes that remain relatively stable under different conditions (Murube et al., 2021).

7.3 Opportunities for machine learning and predictive modeling in trait improvement

However, things are not completely without a turn for the better. Finding patterns in a pile of chaotic data is precisely what machine learning excels at. Nowadays, many research teams are already attempting to use multi-trait prediction models to handle complex traits, such as the yield of common beans and their quality after canning. Especially when genotype and phenotype data are constantly updated, the predictive effects of these models are more reliable than those of traditional methods. Deep learning is also very useful, especially when dealing with high-dimensional and seemingly irregular omics data, as it can help us capture hidden information. Nowadays, an increasing number of candidate genes and metabolites have been screened out precisely through these methods, and they are likely to become key targets in the breeding process in the future (Hanif et al., 2023).

8 Conclusions and Future Perspectives

The quality of a seed is not determined by a single gene. The output, canning performance, protein content and oil level are all supported by a complete set of genetic foundations. Nowadays, the integration of omics data is becoming increasingly in-depth-after the genome, transcriptome, and metabolome are analyzed together, many pathways, key genes, and biomarkers that were previously difficult to see clearly have begun to emerge (Liu et al., 2019). The sequencing of classic strains like B73 and Palomero has opened up a new situation in this field. The new generation of sequencing platforms are not only fast but also have a large volume. Conducting genome-wide association analyses (GWAS) is no longer as challenging as it used to be. More importantly, when you link these big data with trait performance, the predictive model becomes significantly more reliable-especially when genotype and phenotype data are constantly updated. These changes have transformed breeding from "guessing" to "calculating".

However, it is not only technology that is advancing; breeding strategies are also quietly changing. Many decisions that could only be judged based on experience in the past are now being assisted by genomic selection, marker-assisted selection and various predictive models. Multi-omics data is not only seen but also increasingly transformed into part of practical operations. The development of bean varieties has also become more direct and efficient due to these integrated methods. Those hard-to-grasp seed traits are now being gradually pushed to the front line of breeding through selected candidate genes, clear markers and user-friendly databases.

By the way, the challenges in the future may be more complex than they are now, and the tools and ideas needed will not be limited to those of the present. The integration of omics is far from over. The term "pan-omics" is gradually coming into the mainstream view-genomics, transcriptomics, metabolomics, proteomics, epigenetics and phenomics-none of them can fight alone (Zhu et al., 2024). To gain a thorough understanding of seed quality, relying solely on single data is far from sufficient. To make use of these data, it is necessary to rely on tools such as AI, machine learning, and high-throughput phenotyping for assistance (Zhong and Zhong, 2024). Digging out clues from the information and finding directions from the clues is the key to improving the variety in the next step. As for whether precise breeding can be truly achieved and the sustainability of bean production can be enhanced, that depends on the sincerity and execution ability of all parties in future cooperative research, tool development and data sharing.

Acknowledgments

We extend our heartfelt appreciation to Dr. Zhou for her guidance, insightful suggestions, and dedicated contributions during the study’s finalisation.

Conflict of Interest Disclosure

The authors affirm that this research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

References

Aziziaram Z., Cheghamirza K., Zarei L., and Beheshti-Alagha A., 2022, Chemical and morphological characteristics of common bean seed and evaluating genetic advance in commercial classes, Cellular and Molecular Biology, 67(6): 89-99.

https://doi.org/10.14715/cmb/2021.67.6.13

Campbell M., Hu H., Yeats T., Caffe‐Treml M., Gutiérrez L., Smith K., Sorrells M., Gore M., and Jannink J., 2021, Translating insights from the seed metabolome into improved prediction for lipid-composition traits in oat (Avena sativa L.), Genetics, 217(3): iyaa043.

https://doi.org/10.1093/genetics/iyaa043

Cavill R., Jennen D., Kleinjans J., and Briedé J., 2016, Transcriptomic and metabolomic data integration, Briefings in Bioinformatics, 17(5): 891-901.

https://doi.org/10.1093/bib/bbv090

Chaudhary J., Patil G., Sonah H., Deshmukh R., Vuong T., Valliyodan B., and Nguyen H., 2015, Expanding omics resources for improvement of soybean seed composition traits, Frontiers in Plant Science, 6: 1021.

https://doi.org/10.3389/fpls.2015.01021

Chen J., Zeng B., Zhang M., Xie S., Wang G., Hauck A., and Lai J., 2014, Dynamic transcriptome landscape of maize embryo and endosperm development, Plant Physiology, 166(1): 252-264.

https://doi.org/10.1104/pp.114.240689

De Souza L., Scossa F., Proost S., Bitocchi E., Papa R., Tohge T., and Fernie A., 2019, Multi‐tissue integration of transcriptomic and specialized metabolite profiling provides tools for assessing the common bean (Phaseolus vulgaris) metabolome, The Plant Journal, 97: 1132-1153.

https://doi.org/10.1111/tpj.14178

Deng Y., Zheng H., Yan Z., Liao D., Li C., Zhou J., and Liao H., 2018, Full-length transcriptome survey and expression analysis of Cassia obtusifolia to discover putative genes related to aurantio-obtusin biosynthesis, seed formation and development, and stress response, International Journal of Molecular Sciences, 19(9): 2476.

https://doi.org/10.3390/ijms19092476

Di Filippo M., Pescini D., Galuzzi B., Bonanomi M., Gaglio D., Mangano E., Consolandi C., Alberghina L., Vanoni M., and Damiani C., 2021, INTEGRATE: model-based multi-omics data integration to characterize multi-level metabolic regulation, PLoS Computational Biology, 18(2): e1009337.

https://doi.org/10.1371/journal.pcbi.1009337

Dias P., Almeida D., Melo P., Pereira H., and Melo L., 2020, Effectiveness of breeding selection for grain quality in common bean, Crop Science, 61(2): 1127-1140.

https://doi.org/10.1002/csc2.20422

Ding Y., Bi Q., Huang D., Liao J., Yang L., Luo X., Yang P., Li Y., Yao C., Wei W., Zhang J., Li J., Huang Y., and Guo D., 2023, A novel integrated automatic strategy for amino acid composition analysis of seeds from 67 species, Food chemistry, 426: 136670.

https://doi.org/10.1016/j.foodchem.2023.136670

Dossou S., Xu F., You J., Zhou R., Li D., and Wang L., 2021, Widely targeted metabolome profiling of different colored sesame (Sesamum indicum L.) seeds provides new insight into their antioxidant activities, Food Research International, 151: 110850.

https://doi.org/10.1016/j.foodres.2021.110850

Farag M., Khattab A., Maamoun A., Kropf M., and Heiss A., 2019, UPLC-MS metabolome based classification of Lupinus and Lens seeds: a prospect for phyto-equivalency of its different accessions, Food Research International, 115: 379-392.

https://doi.org/10.1016/j.foodres.2018.11.003

Hanif K., Ijaz M., Sajid M., Ammar A., Noor A., and Haris M., 2023, Biotechnological approaches in seed quality enhancement: current status and future prospects, Trends in Animal and Plant Sciences, 1: 37-48.

https://doi.org/10.62324/taps/2023.006

Izquierdo P., Kelly J., Beebe S., and Cichy K., 2023, Combination of meta‐analysis of QTL and GWAS to uncover the genetic architecture of seed yield and seed yield components in common bean, The Plant Genome, 16(2): e20328.

https://doi.org/10.1002/tpg2.20328

Izquierdo P., Wright E., and Cichy K., 2025, GWAS-assisted and multi-trait genomic prediction for improvement of seed yield and canning quality traits in a black bean breeding panel, G3: Genes|Genomes|Genetics, 15(3): jkaf007.

https://doi.org/10.1093/g3journal/jkaf007

Jan S., Rather I., Sofi P., Wani M., Sheikh F., Bhat M., and Mir R., 2021, Characterization of common bean (Phaseolus vulgaris L.) germplasm for morphological and seed nutrient traits from Western Himalayas, Legume Science, 3(2): e86.

https://doi.org/10.1002/leg3.86

Jendoubi T., 2021, Approaches to integrating metabolomics and multi-omics data: a primer, Metabolites, 11(3): 184.

https://doi.org/10.3390/metabo11030184

Kim Y., Oh S., Lee S., Chang A., Park S., Park S., and Kim J., 2024, Metabolic profiling characterizes the genetic diversity of genetically modified and conventional rice, Plant Biotechnology Reports, 18: 863-870.

https://doi.org/10.1007/s11816-024-00940-2

Lei L., Wang L., Wang S., and Wu J., 2020, Marker-trait association analysis of seed traits in accessions of common bean (Phaseolus vulgaris L.) in China, Frontiers in Genetics, 11: 698.

https://doi.org/10.3389/fgene.2020.00698

Li H., Lv Q., Ma C., Qu J., Cai F., Deng J., Huang J., Ran P., Shi T., and Chen Q., 2019, Metabolite profiling and transcriptome analyses provide insights into the flavonoid biosynthesis in the developing seed of tartary buckwheat (Fagopyrum tataricum), Journal of Agricultural and Food Chemistry, 67(40): 11262-11276.

https://doi.org/10.1021/acs.jafc.9b03135

Li L., Hur M., Lee J., Zhou W., Song Z., Ransom N., Demirkale C., Nettleton D., Westgate M., Arendsee Z., Iyer V., Shanks J., Nikolau B., and Wurtele E., 2015, A systems biology approach toward understanding seed composition in soybean, BMC Genomics, 16: S9.

https://doi.org/10.1186/1471-2164-16-S3-S9

Lippolis A., Polo P., De Sousa G., Dechesne A., Pouvreau L., and Trindade L., 2024, High-throughput seed quality analysis in faba bean: leveraging Near-InfraRed spectroscopy (NIRS) data and statistical methods, Food Chemistry: X, 23: 101583.

https://doi.org/10.1016/j.fochx.2024.101583

Manzoor N., Yuan J., Dongcheng W., Liu Z., Lin C., and Mao Z., 2025, Integrated transcriptomic and proteomic analyses revealed molecular mechanisms underlying nutritional changes during seed development of Chenopodium quinoa, Genomics, 117(3): 111045.

https://doi.org/10.1016/j.ygeno.2025.111045

Murube E., Beleggia R., Pacetti D., Nartea A., Frascarelli G., Lanzavecchia G., Bellucci E., Nanni L., Gioia T., Marciello U., Esposito S., Foresi G., Logozzo G., Frega G., Bitocchi E., and Papa R., 2021, Characterization of nutritional quality traits of a common bean germplasm collection, Foods, 10(7): 1572.

https://doi.org/10.3390/foods10071572

Nadeem M., Çilesiz Y., Ali F., Baloch F., and Karaköy T., 2020, Investigation of quality and cooking traits diversity in a global common bean germplasm, Global Journal of Botanical Science, 8: 21-29.

https://doi.org/10.12974/2311-858x.2020.08.3

Niron H., Barlas N., Salih B., and Türet M., 2020, Comparative transcriptome, metabolome, and ionome analysis of two contrasting common bean genotypes in saline conditions, Frontiers in Plant Science, 11: 599501.

https://doi.org/10.3389/fpls.2020.599501

Patt A., Siddiqui J., Zhang B., and Mathé E., 2019, Integration of metabolomics and transcriptomics to identify gene-metabolite relationships specific to phenotype, Methods in Molecular Biology, 1928: 441-468.

https://doi.org/10.1007/978-1-4939-9027-6_23

Pott D., Durán-Soria S., Osorio S., and Vallarino J., 2021, Combining metabolomic and transcriptomic approaches to assess and improve crop quality traits, CABI Agriculture and Bioscience, 2(1): 1-20.

https://doi.org/10.1186/s43170-020-00021-8

Rangan P., Furtado A., and Henry R., 2017, The transcriptome of the developing grain: a resource for understanding seed development and the molecular control of the functional and nutritional properties of wheat, BMC Genomics, 18: 766.

https://doi.org/10.1186/s12864-017-4154-z

Sanches P., De Melo N., Porcari A., and De Carvalho L., 2024, Integrating molecular perspectives: strategies for comprehensive multi-omics integrative data analysis and machine learning applications in transcriptomics, proteomics, and metabolomics, Biology, 13(11): 848.

https://doi.org/10.3390/biology13110848

Sandhu K., You F., Conner R., Balasubramanian P., and Hou A., 2018, Genetic analysis and QTL mapping of the seed hardness trait in a black common bean (Phaseolus vulgaris) recombinant inbred line (RIL) population, Molecular Breeding, 38: 34.

https://doi.org/10.1007/s11032-018-0789-y

Scossa F., Alseekh S., and Fernie A., 2020, Integrating multi-omics data for crop improvement, Journal of Plant Physiology, 257: 153352.

https://doi.org/10.1016/j.jplph.2020.153352

Siddiqui J., Baskin E., Liu M., Cantemir-Stone C., Zhang B., Bonneville R., McElroy J., Coombes K., and Mathé E., 2018, IntLIM: integration using linear models of metabolomics and gene expression data, BMC Bioinformatics, 19: 81.

https://doi.org/10.1186/s12859-018-2085-6

Solanki K., and Abhyankar K., 2024, Metabolomic profiling of germinated and non-germinated Lablab purpureus seeds: antioxidant properties and α-amylase inhibitory activities for diabetes management, Food Production, Processing and Nutrition, 6: 99.

https://doi.org/10.1186/s43014-024-00289-7

Song J., Mavraganis I., Shen W., Yang H., Cram D., Xiang D., Patterson N., and Zou J., 2022, Transcriptome dissection of candidate genes associated with lentil seed quality traits, Plant Biology, 24(5): 815-826.

https://doi.org/10.1111/plb.13426

Wanichthanarak K., Fahrmann J., and Grapov D., 2015, Genomic, proteomic, and metabolomic data integration strategies, Biomarker Insights, 10: 1-6.

https://doi.org/10.4137/BMI.S29511

Yang C., Hu Z., Datla R., and Xiang D., 2025, Editorial: omics in seed development: challenges and opportunities for improving of seed quality and yield in model and crop plants, Frontiers in Plant Science, 16: 1568039.

https://doi.org/10.3389/fpls.2025.1568039

Yi F., Gu W., Chen J., Song N., Gao X., Zhang X., Zhou Y., Ma X., Song W., Zhao H., Esteban E., Pasha A., Provart N., and Lai J., 2019, High temporal-resolution transcriptome landscape of early maize seed development, Plant Cell, 31(5): 974-992.

https://doi.org/10.1105/tpc.18.00961

Yin X., Bose D., Kwon A., Hanks S., Jackson A., Stringham H., Welch R., Oravilahti A., Silva L., Locke A., Fuchsberger C., Service S., Erdos M., Bonnycastle L., Kuusisto J., Stitziel N., Hall I., Morrison J., Ripatti, S., Palotie A., Freimer N., Collins F., Mohlke K., Scott L., Fauman E., Burant C., Boehnke M., Laakso M., and Wen X., 2022, Integrating transcriptomics, metabolomics, and GWAS helps reveal molecular mechanisms for metabolite levels and disease risk, American Journal of Human Genetics, 109: 1727-1741.

https://doi.org/10.1016/j.ajhg.2022.08.007

Yu B., Gao P., Song J., Yang H., Qin L., Yu X., Song H., Coulson J., Bekkaoui Y., Akhov L., Han X., Cram D., Wei Y., Zaharia L., Zou J., Konkin D., Quilichini T., Fobert P., Patterson N., Datla R., and Xiang D., 2023, Spatiotemporal transcriptomics and metabolic profiling provide insights into gene regulatory networks during lentil seed development, The Plant Journal, 115(1): 253-274.

https://doi.org/10.1111/tpj.16205

Zhang H., Hu Z., Yang Y., Liu X., Lv H., Song B., An Y., Li Z., and Zhang D., 2021, Transcriptome profiling reveals the spatial-temporal dynamics of gene expression essential for soybean seed development, BMC Genomics, 22: 453.

https://doi.org/10.1186/s12864-021-07783-z

Zhang Q., Wang S., Qin B., Sun H., Yuan X., Wang Q., Xu J., Yin Z., Du Y., Du J., and Li C., 2023, Analysis of the transcriptome and metabolome reveals phenylpropanoid mechanism in common bean (Phaseolus vulgaris) responding to salt stress at sprout stage, Food and Energy Security, 12(5): e481.

https://doi.org/10.1002/fes3.481

Zhong Y.T., and Zhong S.L., 2024, Development of precision agriculture techniques for soybean yield improvement, Bioscience Evidence, 14(6): 260-269.

https://doi.org/10.5376/be.2024.14.0027

Zhu Z.Y., An X., Luo X.H., Chen C.L., Liu T.T., Zou L.N., and Zhu G.L., 2024, Research progress of NAC transcription factor family in plant stress resistance, Molecular Plant Breeding, 15(3): 90-99.

http://dx.doi.org/10.5376/mpb.2024.15.0011