Cotton Genomics and Genetics 2025, Vol.16, No.5, 249-258 250 genotyping can reshape cotton breeding, thereby fostering superior varieties to address future challenges, and guiding breeders, researchers, and policymakers to utilize molecular tools to achieve sustainable cotton production and global food security. 2 Overview of High-Throughput Genotyping Technologies 2.1 Common platforms and techniques (SNP Arrays, GBS, RAD-seq, DArT, etc.) In fact, high-throughput methods have been attempted for the genotype analysis of cotton for many years. Technologies like SNP chips, GBS, RAD-seq, and DArT, although they have different principles, all have the same goal: to quickly identify those variations related to traits without incurring too much cost. The advantage of chips like CottonSNP63K, CottonSNP80K and ZJU CottonSNP40K lies in the fact that they can "scan" many sites at once and have a high accuracy rate. Especially when conducting intraspecific or interspecific genotyping of cotton, it is relatively convenient to use, mainly relying on the platforms of Illumina or Affymetrix (Shakoor et al., 2017). However, if you are dealing with a highly diverse group or have no idea where the variations are at all, methods like GBS that do not rely on known SNPS would be more appropriate. It relies on next-generation sequencing, testing and searching simultaneously, and is highly flexible. As for RAD-seq and DArT, they also belong to strategies for reducing genomic complexity. Although they are not widely used in cotton, they can still be useful in some situations where resources are limited. 2.2 Technical comparison: throughput, cost, applicability, and data depth Ultimately, different platforms have their own trade-offs. The operation of SNP arrays is simple and the results are reliable, but the SNPS it can detect are pre-designed and cannot "discover" new variations. On the contrary, sequencing-based methods such as GBS are more flexible when the genetic background is complex. However, the analysis process is much more cumbersome than that of chips, and the requirements for data processing are also higher (Cai et al., 2017). 2.3 Data processing pipelines and supporting bioinformatics tools For the subsequent data processing, SNP chips follow a relatively mature set of processes. Manufacturers usually provide ready-made analysis software, and users just need to follow the steps (Hulse-Kemp et al., 2015). But for sequencing technologies like GBS or RAD-seq, the situation is not so "foolproof". You have to do sequence alignment first, then SNP invocation and quality control, and also write some scripts or run open-source tools according to specific projects (Yang et al., 2020). If you still need to proceed with GWAS, QTL mapping or genomic selection, you will have to rely even more on a well-functioning bioinformatics platform to help you integrate a large amount of genotype and phenotype data. 3 Genetic Basis of Key Agronomic Traits in Cotton 3.1 Genetic control of yield-related traits Whether the yield is high or not is not determined by a single factor. Factors like the weight of the bell, the number of bells, and the height of the plant are all quantitative traits, and they are the result of multiple genes working together. Such traits are influenced by additive and dominant genetic effects and are also easily disturbed by environmental changes (Zhang et al., 2020). Current GWAS and QTL studies have identified many candidate gene loci related to these traits. Some regions are quite "capable" and play a role in multiple yield traits (Ma et al., 2024). However, there are still differences among traits. For instance, when it comes to clothing parts, the main gene has a strong control ability and high heritability. However, for instance, the genetic effect of bell numbers is less stable and is easily influenced by multi-gene or environmental interactions (Li et al., 2024). 3.2 Genetic architecture of fiber quality traits When it comes to the indicators that affect fiber quality, the main ones are length, strength and uniformity. However, these traits are not determined by A single gene. Many QTLS are distributed in the A and D subgenomes (Li et al., 2021). Studies have found that QTLS in some gene regions always occur in "clusters", and some act simultaneously on multiple fibrous traits (Huang et al., 2017; Liu et al., 2022; Cai et al., 2024). It is notable that the D subgenome seems to be more active in controlling fiber quality and contains A large number of structural
RkJQdWJsaXNoZXIy MjQ4ODYzNA==