Bt Research 2025, Vol.16, No.4, 136-146 http://microbescipublisher.com/index.php/bt 144 There are also some multiomics databases and platforms specifically targeting microorganisms. For example, the domestic iBioCloud cloud platform provides a module for joint analysis of bacterial genome-tratomes, which can input genome and expression data of Bt and automatically generate functional analysis reports. A typical application of integrated analysis is to combine genomic annotation with transcriptomic data: by mapping transcriptomic differential expression results back to genomic annotation, it is possible to focus on genes that carry both important functions and express significant changes (Kumar et al., 2025). For example, when comparing the transcription of a Bt mutant strain with wild type, it was found that an unknown gene on a plasmid was significantly upregulated, and by genomic annotation and BLAST comparison, it was found that the gene was homologous to the known toxin regulator, which was speculated that the mutation affected a new virulence regulatory pathway. Multiomic integration also includes binding proteomic or metabolomic data (Liu and Zhang, 2024). For example, Zhou et al. conducted whole-genome sequencing and proteome analysis on a Bt strain, corresponding to the toxin sequence identified by the proteome back to the genome-encoded gene, thus confirming that the 13 toxin genes annotated genome were successfully expressed at the protein level. This verification improves the credibility of comments. 8.2 Genomic and omics data visualization tool Effective visualization can transform complex results from Bt genomic and omics data into intuitive information display. For genomic data, common genome browsers such as UCSC Genome Browser and JBrowse can be used to visualize Bt genome and annotate information. Researchers can import the genomic sequence and gene annotation of Bt into the browser, and then superimpose experimental data (such as sequencing reads alignment results, variant sites, expression amounts, etc.) on genomic coordinates in the form of a Track for display. IGV also supports labeling of front and reverse link reads in different colors, which is suitable for the interpretation of prokaryotic polycistronic mRNA. For sequencing results of mutation accumulation experiments, IGV can be used to label each variant on the genome and quickly locate the mutation concentrated region (Reyaz et al., 2019). Another major use of genome browsers is to compare collinearity and variation in genomes of different strains. Heatmap is a common visualization of transcriptome expression display. The expression differences between hundreds of thousands of genes under different conditions of Bt are displayed in a color gradient matrix, so that gene groups with similar expression patterns can be quickly clustered. Volcano plots intuitively show the significance and fold changes of different genes. In the protein-protein interaction network, the Bt protein interaction relationship can be plotted as a network graph using Cytoscape, and the central node of the network (key regulatory protein) can be found. For composite pathways and network results, integrated browsing is also an effective way. 8.3 Application of high-performance computing and cloud platform in large-scale data processing Bt genome research is gradually entering the era of big data, such as population-level comparative genomes, Bt fragment detection in metagenomes, etc., all involve the processing of massive data. To this end, the application of high-performance computing (HPC) and cloud computing platforms has become increasingly important. HPC usually can process the alignment and analysis of hundreds or thousands of Bt genomes in a short time through parallel computing and large storage. For example, the study of comparing the genomes of 90,000 strains of B. cereus group strains has used supercomputing clusters to split and run tasks such as ANI computing in parallel, otherwise it would be difficult for a stand-alone machine to complete such a huge amount of computation. The cloud platform also facilitates team collaboration and data sharing, such as storing the Bt genome annotation and omics data obtained from the analysis in the cloud space. Collaborators at different locations can access and visualize the analysis results at the same time, greatly improving research efficiency. The rise of artificial intelligence and deep learning technologies is also inseparable from HPC support (Wang, 2024). The deep learning framework developed by Hannigan and others predicts biosynthetic gene clusters and requires training models on large GPU servers.
RkJQdWJsaXNoZXIy MjQ4ODYzNA==