Main Page

DNA Microarray

 * Years ago scientists were only able to research few genes during their entire career. This was mainly due to lack of technology and also because one organism has tens of thousands genes. Gene mapping was not only difficult but also time consuming. In late 20th century DNA Microarray was invented which revolutionized the world of genomics, they have the capacity to analyze thousands of genes rapidly and accurately. DNA chip or microarrays uses data mining techniques to analyze genes. There is software in market which can communicate with DNA chip and analyze the data stored in it.



Steps

 * Data mining is a practice of examining large scale database to create meaningful information. Classification, clustering, tree and sequential pattern are the techniques of data mining. These techniques are also being used to analyze DNA microarray. Analyzing microarrays includes several steps like collecting samples, isolating mRNA, creating labeled DNA, hybridization and scanning microarray. Cells are broken, RNA is extracted, the RNA is copied to produce complementary DNA (cDNA), and the cDNA is labeled with fluorescent tags. The cDNA represents genes that are active, that is, being converted to protein via RNA. Converting chip DNA to fluorescent cDNA from untreated and drug-treated cells is called hybridization. A researcher can scan microarrays to simultaneously analyze two different cDNA (or RNA) samples taken from one kind of cell in two different states or from two cell types. To distinguish the samples, each is labeled with a different-colored.

Text Mining Approach

 * There are three stages of microarray data mining pipline, that is, data preprocessing, data modeling or model construction and post processing. These stages are being used by researchers for data mining but it also has a limitation. A new approach to microarray data mining would be text mining™ and information extraction (IE). TM is more concerned with identifying patterns in natural language text and IE is concerned with locating specific entities, relation and facts in text. I will use TM to explain how we can overcome the limitations. Text mining methods create a capable technology for automating the incorporation of scientific knowledge in all the above data mining process. For example, text mining outputs could be combined and correlated with actual gene expression data in model construction such as clustering, classification, and association analysis. Further, text mining can also be applied earlier in data preprocessing for feature selection, data transformation, and data enrichment and in the postprocessing phase for interpretation and knowledge-based validation microarray analyses results.