1 Introduction

Cotton (Gossypium spp.) is a crop that is very important to the world. It not only provides natural fiber, but also produces edible vegetable oil. Therefore, people have always attached great importance to cotton breeding and planting. Now everyone's goal is still the same - hope that cotton can have high and stable yields. However, some difficulties have been encountered in actual planting. For example, the labor force has become less, climate problems have become more serious, and people's requirements for environmental protection and sustainable planting have also become higher (Huang et al., 2022). Recent studies have found that if cotton is to be more drought-resistant and can produce yields in bad environments, its upper and lower structures, especially the underground root system, must be improved (Guo et al., 2024). Only when the roots grow well can they absorb more water and nutrients and survive drought. In addition to breeding technology, the advancement of management methods is also critical. For example, dense planting (multiple sites) and the use of some plant growth regulators have been proven to increase yields and make cotton more suitable for machine harvesting .

When it comes to how cotton looks, plant shape is an important factor. Things like the height of cotton, the length of branches, the angle of branches, and the distribution of leaves will affect how it absorbs sunlight and uses nutrients, and ultimately affect yield (Zhang et al., 2020; An et al., 2022). We can use breeding techniques or planting management methods to adjust the plant type of cotton. This not only allows more cotton to be harvested, but also facilitates machine picking, especially in dense planting (Ji et al., 2020). In fact, when breeding in the past, people have already paid attention to these plant type characteristics. But now we have a new tool - metabolomics. Metabolomics can help us understand the relationship between plant type and yield more clearly. It can reveal which specific metabolites and regulatory pathways are at work behind the scenes. Through this method, we can find out the "key substances" and "regulatory networks" that are helpful to yield. This information can help us more accurately select and cultivate cotton varieties that grow reasonably and have high yields.

This study systematically reviewed the current understanding of how plant architecture affects cotton yield, drawing on the latest advances in genetics, physiology and agronomy, exploring the emerging role of metabolomics in revealing the complex interactions between plant architecture and productivity, and proposing relevant strategies for breeding and management. This study emphasizes the potential of metabolomics in targeted crop improvement. By combining traditional and cutting-edge methods, it can better breed high-yield and stress-resistant cotton varieties, and also help promote the sustainable development of cotton production in the face of future agricultural challenges.

2 Key Factors Regulating Cotton Plant Architecture

2.1 Genotypic variation and architectural traits

The appearance of cotton is related to its genes. Different varieties have different genes, and the height, branch length and angle, and flowering time of cotton will also be different (Figure 1) (Huang et al., 2022). Scientists have found many genes related to these traits through genome-wide association studies. Some genes affect the height and fruit branch length of cotton through auxin signals (Wang et al., 2022). There are also some key genes, such as the PEBP family (such as FT/SFT and SP), which regulate branching patterns and flowering time, which is very helpful for breeding (McGarry and Ayre, 2021). Cotton will behave differently in different regions. For example, in the north, some cotton varieties are more compact and shorter. In other places, some varieties grow taller.

Figure 1  The cultivation feature and plant architecture of cotton (Adopted from Huang et al., 2022) Image caption: (A) The planting pattern of short-dense-early cotton in Xinjiang Uygur Autonomous Region, China. Phenotype of normal fruit branch upland cotton (B) and its schematic diagram (C). Phenotype of short fruit branch upland cotton (D) and its schematic diagram (E). Blue triangles indicate monopodial shoot apical meristem; green triangles represent sympodial shoot meristem; red balls represent determinate floral buds; green peach-like shapes indicate leaves (Adopted from Huang et al., 2022)

2.2 Influence of environmental factors on plant morphogenesis

How cotton grows is not only related to genes, but also has a lot to do with the environment. Factors such as sunlight, temperature, and the length of day and night (photoperiod) will affect plant shape. There is a blue light receptor gene called GhFKF1, which together with another gene GhGI regulates flowering and axillary bud differentiation. Both processes are very sensitive to light and temperature. Some early-maturing cotton varieties will begin to form flowers relatively early. By regulating these genes that are sensitive to the outside world, cotton can be more adaptable to different environments (Li et al., 2024). The balance between limited growth (stopping at a certain extent) and unlimited growth (continuous growth) is also important for cotton. Adjusting this balance can allow cotton to adapt to different planting methods (McGarry and Ayre, 2021).

2.3 Role of endogenous hormones in the establishment of plant architecture

Cotton's own hormones also play a big role in regulating plant shape. For example, there is a regulatory module called miR164-GhCUC2-GhBRC1, which can control the development of lateral buds and the growth of branches by affecting the signal of abscisic acid (ABA). Among them, GhBRC1 can activate the synthesis of ABA, thereby reducing branching (Sun et al., 2021; Zhan et al., 2021). There is also a bHLH transcription factor called GhPAS1, which can regulate brassinosteroids (BR) to promote cell elongation and whole plant growth (Wu et al., 2021). Auxin is also critical. For example, GhHB12 is involved in this signaling pathway, affecting the height and number of branches of cotton (Liu et al., 2022). In addition, gibberellin (GA) is also involved in regulation. For example, the gene GhSBI1 can regulate the length of internodes. It also interacts with DELLA proteins to affect cell elongation and GA signaling (Zhong et al., 2024). These hormones will influence each other, integrate genetic and environmental signals together, and finely regulate the structure of cotton. In this way, cotton can grow high yields and adapt to different environments.

3 Advances in Metabolomics Research in Cotton

3.1 Establishment and integration of cotton metabolomics databases

In recent years, there are more and more metabolomics resources for cotton. Researchers have collected metabolic data at different developmental stages and different parts, and also integrated transcriptome data. For example, a platform called Cotton Metabolism Regulatory Network (CMRN) was established in the study, which contains more than 2 100 metabolites and more than 90 000 genes. These data have provided great help for everyone to study the growth, structure and yield of cotton (Liu et al., 2024). There are also some databases and visualization tools that specifically collect information on cotton ovules and fiber development. These tools make it easier for researchers to share data and do comparative analysis.

3.2 Association analysis of key metabolic pathways and high-yield traits

Scientists analyzed the metabolome and transcriptome data together and found that some metabolic pathways are particularly related to the high-yield traits of cotton. For example, pathways such as phenylpropanoid synthesis, tyrosine metabolism, and phenylalanine metabolism are related to cotton bud differentiation and flowering time. A gene called GhTYDC-A01 was found to affect these processes. In the process of cotton fiber development, another class of substances called very long chain fatty acids (VLCFA) is also critical. They are synthesized through the fatty acid elongation pathway. And genes like GhKCS1b_Dt are important regulators in this process (Liu et al., 2024). In addition, during the somatic embryonic development of cotton, scientists found that purine metabolism and flavonoid synthesis are also changing. This suggests that they may affect how cells differentiate and regenerate (Guo et al., 2019).

3.3 Mining and functional validation of targeted metabolites

Now, through metabolomics, scientists can more accurately find which metabolites affect cotton growth, stress resistance, and yield. Different accumulation levels of substances such as phenolic acids, flavonoids, and amino acids will affect the early maturity of cotton, fiber quality, and response to environmental stress (Han et al., 2023; Liu et al., 2024). The researchers also verified the function of genes. For example, after overexpressing or silencing candidate genes such as GhTYDC-A01 or GhKCS1b_Dt, it was found that they can really affect flowering time and fiber elongation. These research results show that we can optimize the structure of cotton and increase yield through the method of "targeted metabolic regulation". This is a very promising path for breeding and metabolic engineering.

4 Metabolomics-Driven Strategies for Optimizing Plant Architecture

4.1 Screening for ideal plant types based on metabolic characteristics

Now, researchers can use large-scale metabolite analysis methods to find metabolic characteristics related to good cotton structure and yield (Hong et al., 2016). If the metabolome data is combined with genetics and transcriptome, key genes can be located more accurately. In this way, subsequent functional verification and screening can also be done faster (Kumar et al., 2017; Shen et al., 2022). This method helps us find cotton varieties with suitable plant types and good metabolic performance more quickly. Especially in breeding, it can save a lot of time and energy.

4.2 Early prediction of plant architecture using metabolic markers

The metabolome can also be used for prediction. It can identify metabolic markers related to plant structure, which can predict future performance before the cotton grows (Villate et al., 2020). These metabolic markers are like "early warning tools" that can pick out genotypes with great potential before the traits are expressed. This is particularly useful for breeding. Moreover, if metabolomics is used together with other "omics" platforms, the prediction results will be more accurate (Kumar et al., 2017; Shen et al., 2022). This will make the improvement work more targeted and more efficient.

4.3 Directed improvement of architecture through metabolic pathway regulation

To purposefully change how cotton grows, we must first understand the metabolic pathways that control its plant type (Hong et al., 2016; Shen et al., 2022). With this information, we can start through metabolomics breeding or metabolic engineering. For example, adjust the pathways that control plant hormone synthesis or stress resistance to make cotton grow the way we want (Kumar et al., 2017). This method does not rely on luck, but directly regulates the biochemical reactions that occur in plants. This can more accurately improve the structure of cotton, increase its yield, resistance and ability to adapt to the environment (Raza, 2020).

5 Identification of Yield-Related Metabolites

5.1 Carbon and nitrogen metabolites associated with reproductive growth

Some key carbon and nitrogen metabolites, such as sucrose, alanine, aspartic acid, citric acid and malic acid, are closely related to cotton reproductive growth and yield (Levi et al., 2011; Jiang et al., 2012). If the activity of sucrose synthase (GhSusA1) is enhanced, the fiber yield and biomass of cotton can be increased. At the same time, if the nitrogen content in the leaves increases, it can also increase the activity of the enzyme, enhance photosynthesis, and make more sucrose. This series of changes will eventually make the cotton bolls heavier and the seeds more (Iqbal et al., 2022). During drought, the levels of some amino acids and organic acids in the plant will increase. This change can help cotton regulate its body water, alleviate the effects of drought, and also help restore reproductive growth.

5.2 Secondary metabolites involved in biomass accumulation and distribution

Cotton also produces some secondary metabolites, such as phenolic acids, flavonoids, and compounds involved in the metabolism of phenylpropanoids, tyrosine, and phenylalanine. These substances are particularly important when cotton copes with adversities (such as pests and diseases, drought, strong light, etc.). They can help plants enhance their defense capabilities and make cotton yield more stable under difficult environments (Prakash et al., 2023). These metabolic pathways are also related to the early maturity and fiber quality of cotton. By regulating these metabolic processes, the overall performance of cotton may also be improved.

5.3 Comparative metabolomic profiles of high-yield cotton lines

The metabolome comparison of some high-yield cotton lines found that their metabolites had some obvious characteristics. For example, these varieties have higher levels of certain solutes and metabolites. High-yield cotton may use different physiological strategies to increase yield. For example, they may use light more efficiently, have a higher harvest index, or accumulate more biomass (Virk et al., 2023). In near-isogenic lines with yield-related QTLs introduced, it was found that these cottons accumulated more glycerol, inositol, and some organic acids under stressful conditions. The increase in these metabolites may be related to their high yield and enhanced stress resistance (Levi et al., 2011). These research results show that metabolomics can help us find out which metabolites are related to yield. This is very valuable for screening and breeding high-yield cotton varieties.

6 Integration of Metabolomics with Genetic Breeding

6.1 Combined analysis of metabolomics and QTLs

Nowadays, many studies will combine metabolomics data with QTL positioning. This can find some genetic regions called mQTL, which are gene locations related to certain metabolites (Litvinov et al., 2021; Sakurai, 2022). In this way, we can link the changes in a certain metabolite with the genes behind it. In this way, it is easier to find the key pathways that affect cotton structure and yield, or important "candidate genes" (Fernie and Schauer, 2009). In addition, analyzing metabolites and QTLs together can also improve the accuracy of molecular markers used in breeding, and can also more quickly introduce useful genes into excellent cotton varieties.

6.2 Multi-omics integration to improve breeding efficiency

Metabolomics is not a one-man battle, it can also be used in conjunction with genomics, transcriptomics and proteomics. These omics data combined can help us understand the relationship between "genotype" and "traits" from more perspectives (Scossa et al., 2020). For example, which genes determine yield? Which are related to disease resistance and drought tolerance? These can be seen more clearly. Through these combined analyses, we can also find out more accurately "which gene is the problem", figure out which metabolic pathway is at work, and know how genes are regulated (Sharma et al., 2021). This method can also be combined with some high-throughput analysis tools to provide more clues for gene editing and breeding strategies, helping us develop better cotton varieties (Razzaq et al., 2022).

6.3 Construction of metabolomics-driven molecular breeding models

Current research has also developed some metabolome-based breeding models. These models incorporate metabolic markers and pathway information to guide how to select ideal traits (Razzaq et al., 2022). In this way, we can screen out some cotton lines with great potential in the early stages of breeding. For example, varieties that are more resistant to stress, higher in yield, and more adaptable to climate change (Sakurai, 2022). Combining metabolomics and bioinformatics tools with some rapid breeding technologies can also speed up the development of new varieties. This is also particularly helpful for us to cope with various environmental challenges (Razzaq et al., 2019).

7 Case Studies

7.1 Improvement of compact cotton varieties based on metabolic traits

Some studies have used metabolomics and transcriptomics to compare different cotton varieties, such as Zhongmian 50 (early maturing, compact) and Guoxin Cotton 11 (late maturing). The results showed that several metabolic pathways are particularly important, such as phenylpropanoid synthesis, tyrosine metabolism, and phenylalanine metabolism, which are all related to bud differentiation and early flowering. The study also noted that the content of phenolic acids decreases in early-maturing cotton. A gene called GhTYDC-A01 was also found to be related to flowering time. These results suggest that we can select more compact and faster-maturing cotton varieties by regulating metabolites and genes.

7.2 Functional validation of specific metabolites in regulating flower and boll distribution

Functional studies have also found some specific metabolites and genes that are important for the development of cotton. For example, a study overexpressed the gene GhTYDC-A01 in Arabidopsis thaliana and found that the flowering time of these plants was delayed. This shows that this gene can indeed affect flowering time and is an important factor in regulating reproductive development. In addition, spatial metabolomics also helped to find several key metabolites that affect cotton fiber formation, such as linoleic acid, spermine and spermidine. Through gene knockdown or overexpression experiments, researchers confirmed that these metabolites play an important role in fiber cell initiation and development.

7.3 Pilot study on large-scale field screening using metabolomics data

Metabolomics is now also beginning to be applied in field breeding. Researchers conducted a pilot study combining spatial metabolomics and lipidomics to analyze many different cottonseed varieties. They found 17 differential metabolites and 125 lipids related to stress resistance. These lipids can help plants remove reactive oxygen species and regulate osmotic pressure, which is very helpful for improving drought and salt tolerance (Liu et al., 2021). This shows that it is feasible to use metabolomics technology for screening in the field. We can select cotton varieties with good structure, high yield and strong stress resistance more quickly.

8 Concluding Remarks

This study highlights the importance of metabolomics. It can help us find out which metabolites and metabolic pathways affect cotton plant type and yield. When we analyze metabolomics data together with genetic information and transcriptome data, we can find metabolic markers related to high yield, stress resistance, and ideal plant type. These findings allow us to judge the performance of varieties earlier, and to carry out breeding improvements more targetedly, accelerating the process of breeding high-performance cotton.

However, there are still many difficulties in using metabolomics in cotton breeding. First of all, the genome of cotton is very complex, it is polyploid, and there are many duplicated genes. This makes it difficult for us to find those truly useful metabolic genes at once. In addition, we still lack a standardized metabolic database specifically prepared for cotton. Many of the data used now are scattered and not very unified. Another point is that the metabolome and other "omics" data (such as genome and proteome) have not been integrated well. If these data can be combined better, we can see more comprehensively how various traits come from. In addition, if you want to use a good method developed in the laboratory in the field, you have to try it several times. Not only do you need to improve the method itself, but you also need to verify it repeatedly under different environments to know whether it is stable and reliable.

Although there are still challenges, metabolomics still has great prospects in cotton breeding. Today's high-throughput analysis technology is getting faster and faster, and the integration technology between different "omics" is becoming more and more mature. These advances can help us save time and improve efficiency in breeding. In the future, we can combine metabolome data with genome and transcriptome data, and use gene editing tools such as CRISPR/Cas to more quickly identify key pathways that control complex traits. In this way, we can build a metabolome-driven breeding model, and then use some rapid breeding methods to breed cotton varieties that are adaptable to climate change and have high yields. With these means, cotton production can also go more steadily and further in the face of global climate change.

Acknowledgments

We thank Mr Z. Tao from the Institute of Life Science of Jiyang College of Zhejiang A&F University for his reading and revising suggestion.

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

An J., Zhang Z., Li X., Xing F., Lei Y., Yang B., Wang Z., Han Y., Chen H., Wang G., Feng L., Du W., and Li Y., 2022, Loose and tower-type canopy structure can improve cotton yield in the Yellow River basin of China by increasing light interception, Archives of Agronomy and Soil Science, 69(6): 920-933.

https://doi.org/10.1080/03650340.2022.2045584

Fernie A., and Schauer N., 2009, Metabolomics-assisted breeding: a viable option for crop improvement? Trends in Genetics, 25(1): 39-48.

https://doi.org/10.1016/j.tig.2008.10.010

Guo C., Bao X., Sun H., Zhu L., Zhang Y., Zhang K., Bai Z., Zhu J., Liu X., Li A., Dong H., Zhan L., Liu L., and Li C., 2024, Optimizing root system architecture to improve cotton drought tolerance and minimize yield loss during mild drought stress, Field Crops Research, 308: 109305.

https://doi.org/10.1016/j.fcr.2024.109305

Guo H., Guo H., Zhang L., Tang Z., Yu X., Wu J., and Zeng F., 2019, Metabolome and transcriptome association analysis reveals dynamic regulation of purine metabolism and flavonoid synthesis in transdifferentiation during somatic embryogenesis in cotton, International Journal of Molecular Sciences, 20(9): 2070.

https://doi.org/10.3390/ijms20092070

Han M., Cui R., Wang D., Huang H., Rui C., Malik W., Wang J., Zhang H., Xu N., Liu X., Lei Y., Jiang T., Sun L., Ni K., Fan Y., Zhang Y., Wang J., Chen X., Lu X., Yin Z., Wang S., Guo L., Zhao L., Chen C., and Ye W., 2023, Combined transcriptomic and metabolomic analyses elucidate key salt-responsive biomarkers to regulate salt tolerance in cotton, BMC Plant Biology, 23(1): 245.

https://doi.org/10.1186/s12870-023-04258-z

Hong J., Yang L., Zhang D., and Shi J., 2016, Plant metabolomics: an indispensable system biology tool for plant science, International Journal of Molecular Sciences, 17(6): 767.

https://doi.org/10.3390/ijms17060767

Huang X., Liu H., and Ma B., 2022, The current progresses in the genes and networks regulating cotton plant architecture, Frontiers in Plant Science, 13: 882583.

https://doi.org/10.3389/fpls.2022.882583

Iqbal A., Jing N., Qiang D., Wang X., Gui H., Zhang H., Pang N., Zhang X., and Song M., 2022, Physiological characteristics of cotton subtending leaf are associated with yield in contrasting nitrogen-efficient cotton genotypes, Frontiers in Plant Science, 13: 825116.

https://doi.org/10.3389/fpls.2022.825116

Ji G., Liang C., Cai Y., Pan Z., Meng Z., Li Y., Jia Y., Miao Y., Pei X., Gong W., Wang X., Gao Q., Peng Z., Wang L., Sun J., Geng X., Wang P., Chen B., Wang P., Zhu T., He S., Zhang R., and Du X., 2020, A copy number variant at the HPDA-D12 locus confers compact plant architecture in cotton, New Phytologist, 229(4): 2091-2103.

https://doi.org/10.1111/nph.17059

Jiang Y., Guo W., Zhu H., Ruan Y., and Zhang T., 2012, Overexpression of GhSusA1 increases plant biomass and improves cotton fiber yield and quality, Plant Biotechnology Journal, 10(3): 301-312.

https://doi.org/10.1111/j.1467-7652.2011.00662.x

Kumar R., Bohra A., Pandey A., Pandey M., and Kumar A., 2017, Metabolomics for plant improvement: status and prospects, Frontiers in Plant Science, 8: 1302.

https://doi.org/10.3389/fpls.2017.01302

Levi A., Paterson A., Cakmak I., and Saranga Y., 2011, Metabolite and mineral analyses of cotton near-isogenic lines introgressed with QTLs for productivity and drought-related traits, Physiologia Plantarum, 141(3): 265-275.

https://doi.org/10.1111/j.1399-3054.2010.01438.x

Li X., Wu Y., Liu Z., Wei H., Wang H., and Yu S., 2024, Apical meristem transcriptome analysis identifies a role for the blue light receptor gene GhFKF1 in cotton architecture development, The Crop Journal, 12(4): 1126-1136.

https://doi.org/10.1016/j.cj.2024.06.009

Litvinov D., Karlov G., and Divashuk M., 2021, Metabolomics for crop breeding: general considerations, Genes, 12(10): 1602.

https://doi.org/10.3390/genes12101602

Liu B., Wang X., Li K., and Cai Z., 2021, Spatially resolved metabolomics and lipidomics reveal salinity and drought-tolerant mechanisms of cottonseeds, Journal of Agricultural and Food Chemistry, 69(28): 8028-8037.

https://doi.org/10.1021/acs.jafc.1c01598

Liu Y., Wang W., Zhang L., Zhu L., Zhang X., and He X., 2022, The HD-Zip transcription factor GhHB12 represses plant height by regulating the auxin signaling in cotton, Journal of Integrative Agriculture, 22(7): 2015-2024.

https://doi.org/10.1016/j.jia.2022.09.022

Liu Z., Fan L., Shu S., Qanmber G., Chen E., Huang J., Li F., and Yang Z., 2024, Cotton metabolism regulatory network: Unraveling key genes and pathways in fiber development and growth regulation, Plant Communications, 6(3): 101221.

https://doi.org/10.1016/j.xplc.2024.101221

McGarry R., and Ayre B., 2021, Cotton architecture: examining the roles of SINGLE FLOWER TRUSS and SELF-PRUNING in regulating growth habits of a woody perennial crop, Current Opinion in Plant Biology, 59: 101968.

https://doi.org/10.1016/j.pbi.2020.10.001

Prakash S., Kumar M., Ma R., Kumar S., Jaconis S., Parameswari E., Sharma K., Dhumal S., Senapathy M., Deshmukh V., Dey A., Lorenzo J., Sheri V., and Zhang B., 2023, The resilient cotton plant: uncovering the effects of stresses on secondary metabolomics and its underlying molecular mechanisms, Functional & Integrative Genomics, 23(2): 183.

https://doi.org/10.1007/s10142-023-01118-9

Raza A., 2020, Metabolomics: a systems biology approach for enhancing heat stress tolerance in plants, Plant Cell Reports, 41(3): 741-763.

https://doi.org/10.1007/s00299-020-02635-8

Razzaq A., Sadia B., Raza A., Hameed M., and Saleem F., 2019, Metabolomics: a way forward for crop improvement, Metabolites, 9(12): 303.

https://doi.org/10.3390/metabo9120303

Razzaq A., Wishart D., Wani S., Hameed M., Mubin M., and Saleem F., 2022, Advances in metabolomics-driven diagnostic breeding and crop improvement, Metabolites, 12(6): 511.

https://doi.org/10.3390/metabo12060511

Sakurai N., 2022, Recent applications of metabolomics in plant breeding, Breeding Science, 72(1): 56-65.

https://doi.org/10.1270/jsbbs.21065

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

Sharma V., Gupta P., Priscilla K., Sharankumar S., Hangargi B., Veershetty A., Ramrao D., Suresh S., Narasanna R., Naik G., Kumar A., Guo B., Zhuang W., Varshney R., Pandey M., and Kumar R., 2021, Metabolomics intervention towards better understanding of plant traits, Cells, 10(2): 346.

https://doi.org/10.3390/cells10020346

Shen S., Zhan C., Yang C., Fernie A., and Luo J., 2022, Metabolomics-centered mining of plant metabolic diversity and function: past decade and future perspectives, Molecular Plant, 16(1): 43-63.

https://doi.org/10.1016/j.molp.2022.09.007

Sun Q., Xie Y., Li H., Liu J., Geng R., Wang P., Chu Z., Chang Y., Li G., Zhang X., Yuan Y., and Cai Y., 2021, Cotton GhBRC1 regulates branching, flowering, and growth by integrating multiple hormone pathways, The Crop Journal, 10(1): 75-87.

https://doi.org/10.1016/J.CJ.2021.01.007

Villate A., Nicolas S., Gallastegi M., Aulas P., Olivares M., Usobiaga A., Etxebarria N., and Aizpurua‐Olaizola O., 2020, Metabolomics as a prediction tool for plants performance under environmental stress, Plant Science, 303: 110789.

https://doi.org/10.1016/j.plantsci.2020.110789

Virk G., Snider J., Chee P., Virk S., Parkash V., Kaur N., and Jones D., 2023, Physiological contributors to yield in advanced cotton breeding lines, Crop Science, 63(6): 3264-3276.

https://doi.org/10.1002/csc2.21082

Wang C., Ma Q., Xie X., Zhang X., Yang D., Ma X., Lin H., and Su J., 2022, Identification of favorable haplotypes/alleles and candidate genes for three plant architecture-related traits via a restricted two-stage multilocus genome-wide association study in upland cotton, Industrial Crops and Products, 177: 114458.

https://doi.org/10.1016/j.indcrop.2021.114458

Wu H., Ren Z., Zheng L., Guo M., Yang J., Hou L., Qanmber G., Li F., and Yang Z., 2021, The bHLH transcription factor GhPAS1 mediates BR signaling to regulate plant development and architecture in cotton, The Crop Journal, 9(5): 1049-1059.

https://doi.org/10.1016/j.cj.2020.10.014

Zhan J., Chu Y., Wang Y., Diao Y., Zhao Y., Liu L., Wei X., Meng Y., Li F., and Ge X., 2021, The miR164‐GhCUC2‐GhBRC1 module regulates plant architecture through abscisic acid in cotton, Plant Biotechnology Journal, 19(9): 1839-1851.

https://doi.org/10.1111/pbi.13599

Zhang X., Rui Q., Li Y., Chen Y., Zhang X., Chen D., and Song M., 2020, Architecture of stem and branch affects yield formation in short season cotton, Journal of Integrative Agriculture, 19(3): 680-689.

https://doi.org/10.1016/s2095-3119(19)62626-2

Zhong W., Wu L., Li Y., Li X., Wang J., Pan J., Zhu S., Fang S., Yao J., Zhang Y., and Chen W., 2024, GhSBI1, a CUP‐SHAPED COTYLEDON 2 homologue, modulates branch internode elongation in cotton, Plant Biotechnology Journal, 22(11): 3175-3193.

https://doi.org/10.1111/pbi.14439