Computational Molecular Biology 2025, Vol.15, No.6, 299-306 http://bioscipublisher.com/index.php/cmb 302 Figure 1 Example output for the flanking regions of an identified HGTs (Adopted from Song et al., 2019) 5 Case Study: Computational Detection of HGT in an Agricultural Soil Microbiome 5.1 Study background: sampling agricultural soils under conventional and organic practices Before conducting HGT analysis, how and from where the soil is collected actually largely determine what can be observed subsequently. The samples used in this study were from farmlands with different farming methods, including traditional, organic, and plots that had been fertilized with manure or irrigated with sewage. These management measures often change the local microbial composition and naturally affect the frequency and mode of HGT (Ren et al., 2022). Researchers pay particular attention to antibiotic resistance genes (ARGs) and mobile genetic elements (MGEs) mainly because they are more closely associated with ecological risks and health problems. The sampling strategy is not merely for comparing the farming methods themselves, but aims to present the changing patterns of HGT under different agricultural disturbances as much as possible. 5.2 Implementation of HGT detection pipeline using metagenomic assemblies and alignment tools When processing these soil data, the research team did not immediately conduct HGT determination. Instead, they first assembled and bogged the metagenomic data, trying to piece together the genomes that could be reconstructed as much as possible. Only after the sequence framework is clear can the method of comparative genomics and phylogenetic inconsistency be used to determine which genes may be "foreign" (Wijaya et al., 2025). Meanwhile, researchers have also incorporated some ideas of statistical models and machine learning, such as taking into account the differences in gene length and alignment biases caused by closely related species (Figure 2) (Sevillya et al., 2020). These methods do not replace each other but are combined into the same process to ensure that the HGT signals in complex agricultural soils are not masked by disordered data. 5.3 Major findings: prevalent mobile gene families, microbial donors/recipients, and environmental influences When analyzing the results, a phenomenon that is hard to ignore is that in the soil that has been treated with manure, there are significantly more mobile gene families related to resistance and metabolism, especially shortly after application, when gene exchanges seem to be more frequent. However, the sources of mobile genes are not singular. The study simultaneously identified multiple possible donors and recipients, including the common Bacillus and Nocardia genera in soil, as well as the Comonas genus from fecal backgrounds. This indicates that genes can interweave and flow among different ecological sources. It is worth noting that, apart from manure, factors such as sewage irrigation and pollution from mining can also alter the frequency of HGT and the way drug resistance genes spread. Overall, the impact of agricultural management measures on the gene transfer network is more direct than imagined.
RkJQdWJsaXNoZXIy MjQ4ODYzNA==