core gene selection through different approaches
Characterizing cellular states in response to various disease conditions is an important issue which is addressed by di erent methods such as Large-scale gene expression pro ling. One of the most important challenges in front of bioinformaticians is the loss of data because expression pro ling is still very expensive. It is understood that pro ling a group of selected genes could be enough for understanding all of the gene expression pro le. In this research, we propose a fast method for estimation of the missing values in low-rank matrices. We consider the highly correlated expression pro les as a low-rank matrix. Then, we used this new method in a proposed algorithm which will select the landmark genes and also estimate the target genes iteratively. The algorithm tries to enhance the representation of the landmark genes in each iteration.The proposed algorithm was successful compared to the related works. It could help in the process of lowering the expenses of the Large-scale gene-expression pro ling. The improvement of the estimation error was 3.2 percent compare to the best method which is D-GEX. Also we capture the errors which are signi cant in both methods and nd that we could reduce the signi cant errors 21 percent in comparison with D-GEX.