data preprocessing

Achieving technologies that can depict the behaviors of genes at the transcriptome level and the interaction between them is particularly beneficial. In this regard, the microarray technology is based on the previous discoveries of qPCR and hybridization blotting technology, which uses the same rules but on a larger scale to measure the expression of several thousand genes. In fact, it is a method based on hybridization in combination with nanotechnology and the use of labeled fluorescent dyes. This technology is based on a chip and a glass plate, where special identifier sequences are placed together in a two-dimensional order of several tens of thousands of units, and it's responsible for simultaneously identifying the differential expression pattern of several thousand special sequences or coding regions. Prior knowledge of the microarray architecture and backstage details used for this technique has a direct impact on raw data analysis. Data pre-processing methods and their normalization, using the principal component analysis method, Heatmap drawing, clustering, and data correlation test dendrogram drawing, and the advanced TSN method are among the leading methods. Also, the challenge of comparing multiple averages, the existence of type one error, and the concept of fold change are also briefly discussed. In the present study, it has tried to discuss the theoretical foundations, raw data structure, statistical analysis, and interpretation of the results of microarray technology and to be able to independently analyze the path of data to interpretation as well as overcome the existing challenges with the information that is given.