Plant MicroRNA Gene Prediction Preliminary Study

The microRNA (miRNA) gene was first discovered in1993. After being obscurity foralmost a decade, this gene is recognized as the founding member of a new class of regulatoryRNAs now. The microRNA genes express~22nucleotide (nt) RNAs which can regulate theexpression of protein-coding genes with sequences of antisense complementarity. ThemicroRNA has played a central role in a series of vital life process, including earlydevelopment, cell proliferation, fat metabolism and cell death.The intense interest inunderstanding the role of miRNAs in regulating gene expression has fueled the developmentof new methods to study how these tiny RNA genes are expressed and function. ThemicroRNAs derive from~70-90nucleotides single-stranded RNA precursor with hairpinstructure by the Dicer enzyme both in animal and plant.In contrast to animal, plant microRNAmainly act as small interfering RNAs(siRNAs) guiding the destruction of mRNA target.The search for plant miRNA in contrast to plant, the mode of action of plant miRNAsdetermines its degree of integration with mRNA-complete combination. Though there exists acertain degree of common ground in feature extraction methods, but the characteristics ofplant miRNA in some of the features set the two types of miRNA prediction apart.Characteristics of plant miRNA are not as complex as animals. When forecasting, it wouldrequire a more precise algorithm or a more appropriate model to guide the prediction.The core factors for feature extraction plant miRNA, falls mainly on the following aspects:(1)the conservation of mature plant miRNA;(2)binding-sites;(3)stability of plant miRNAprecursor secondary structure;(4) probability distribution of minimum free energy of theplant miRNA precursor;(5)the cut length of the plant miRNA precursor site. According tothe feature extraction methods above, the researchers have developed a series of plant miRNAprediction tool. These prediction tools are usually based on Smith-Waterman algorithm to bebuilt up, such as PatScan, miRU, psRNATarget and so on. Especially for the construction ofmiRDeep on the model, use optimization algorithms to optimize the model parameters, andmake it more suitable for plant miRNA predictions.If some gene can transcript mRNAs, which can express curcin and make host（plant oranimal）ill, who may targeted by these tiny RNA generated by endogenous at varying levels, that means, miRNA will take considerable income. Corncerning the farm crop among botany,the most obvious advantages are increasing production, preventing lodging, resistanting pest,etc. For the purspose of supplying convenience to the scientific researchers, we summarize thecorresponding methods of seeking trait extraction of plant miRNAs and the related softwaresin this paper. According to the learning of above softwares, I have some awareness about plantmiRNA prediction, this paper achieve three base pairing characteristics and be able to predictthe plant miRNA.