PROJECT TITLE :
A PSO-Based Approach for Pathway Marker Identification From Gene Expression Data
In this text, a replacement and strong pathway activity inference theme is proposed from gene expression data using Particle Swarm Optimization (PSO). From microarray gene expression data, the corresponding pathway information of the genes are collected from a public database. For identifying the pathway markers, the expression values of each pathway consisting of genes, termed as pathway activity, are summarized. To measure the goodness of a pathway activity vector, t-score is widely used in the present literature. The weakness of existing techniques for inferring pathway activity is that they intend to think about all the member genes of a pathway. But really, all the member genes may not be significant to the corresponding pathway. Therefore, those genes, that are accountable within the corresponding pathway, should be included solely. Motivated by this, in the proposed method, using PSO, important genes with respect to every pathway are identified. The target is to maximise the common t-score. For the pathway activities inferred from totally different percentage of vital pathways, the common absolute t-scores are plotted. Yet, the prime fiftyp.c pathway markers are evaluated using 10-fold cross validation and its performance is compared with that of other existing techniques. Biological relevance of the results is additionally studied.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here