Computational Molecular Biology 2025, Vol.15, No.5, 227-234 http://bioscipublisher.com/index.php/cmb 229 adding biological repetition groups, although these operations are cumbersome, are to ensure that drought-responsive mirnas are not overlooked. 3.2 Data filtering, miRNA identification, and annotation methods The raw data obtained cannot be directly analyzed. The first step must be to "clean it up" first. Sequences like adaptor sequences, low-quality reads, and small molecules that are clearly not mirnas (such as tRNA, rRNA, snoRNA) will all be filtered out. The remaining high-quality sRNA sequences are then compared with the reference genome. If one is lucky, they can find conserved mirnas that have been recorded in databases such as miRBase. When encountering the unknown, researchers have to resort to tools. Software such as miRDeep2, miRA or miRDeepFinder can predict new miRNA candidates based on sequence abundance, precursor structure and secondary structure (Evers et al., 2015). Sometimes, these predictions still need to be further confirmed, such as whether the miRNA precursors have typical stem-loop structures, whether the sequences are conserved, and how the expression levels vary among different samples. As for exactly which mrnas they target, tools such as psRNATarget and CleaveLand come in handy, and degradation omics sequencing is often required for verification (Xie et al., 2012; Sepulveda-Garcia et al., 2020). 3.3 Identification of differentially expressed miRNAs between drought-treated and control samples Not every miRNA responds during drought; those with significant changes in expression are the focus of researchers. To identify these "responders", statistical tools such as DESeq2 are needed to compare the expression levels of mirnas in the drought group and the control group one by one (Sharma et al., 2025). In the research on corn, many mirnas showed significant up-regulation or down-regulation under drought conditions. Some were familiar faces, while new discoveries were made (Aravind et al., 2017; Liu et al., 2019). However, sequencing alone is not enough. qRT-PCR or Northern blotting is usually used as verification methods to confirm whether these differential expressions truly exist. This step is crucial because many of the miRNA-mRNA regulatory modules to be analyzed subsequently have been screened out from this batch of differentially expressed mirnas. 4 Functional Prediction and Analysis of Drought-Responsive miRNAs 4.1 Target gene prediction methods and bioinformatics tools To figure out exactly what role a miRNA plays in drought response, the most direct way is to see who it regulates. Target gene prediction may sound highly technical, but in fact, the principle is not complicated - it relies on sequence complementarity and the accessibility of binding sites. Like in Corn, tools such as psRNATarget, psRobot and TargetFinder have become the "old three" that everyone commonly uses (Tang et al., 2022). However, predictions are predictions, but they cannot be implemented. Therefore, many times researchers will bring in degradation group sequencing data to cross-verify whether miRNA is indeed functioning (Yang et al., 2025). In addition, some people are more cautious and simply incorporate multiple sequence alignment (Clustal Omega), cluster analysis (R packages like seqinR and ape), and co-expression networks to help confirm whether the regulatory relationship under drought is "reliable". This is like a jigsaw puzzle. Only when each piece is pieced together can a reliable regulatory map be formed. 4.2 GO annotation and KEGG pathway analysis for functional enrichment The predicted targets cannot merely be put up on a list; they must be explained exactly what they do during droughts. At this point, GO and KEGG come in handy. GO annotations categorize these genes into large boxes such as "Molecular functions" and "biological processes". Some common entries include cellular response regulation, water stress response, etc. (Liu et al., 2019; Jiao et al., 2022). KEGG, on the other hand, is more pastration-oriented and will tell you which signaling pathways are regulated by miRNA. Classic drought-resistant pathways such as plant hormone transmission, glutathione metabolism, and phenylpropanin synthesis are often on the list. To conduct these analyses, Blast2GO and AgriGO are frequently used tools. 4.3 Identification of key miRNAs involved in drought stress regulation Some mirnas "step up" during drought and become very active, especially some classic ones such as miR164, miR159, miR156, miR319, miR160, as well as miR394 and miR408a, which are almost "familiar faces" in corn. They do not work on their own but "command the battle" by regulating transcription factors, such as MYB, NAC,
RkJQdWJsaXNoZXIy MjQ4ODYzNA==