Predicting drug-resistant miRNAs in cancer

AbstractMicroRNAs (miRNAs) are small noncoding RNAs that play a vital role in controlling drug sensitivity/resistance in cancer. Hence, the effectiveness of cancer treatment can be significantly improved by identifying these miRNAs. We reviewed the studies that emphasize the identification of miRNAs associated with drug resistance in cancer. Two computational methods are also developed to identify the miRNAs associated with drug resistance. In the first method, the Euclidean distance with weighted fold change (EDWFC) score is developed. Here, the Euclidean distance and fold change, computed using the averages of control and resistant expressions, are multiplied by varying the power of fold change to minimize the average rank of miRNAs. The miRNAs are then sorted in descending order according to the EDWFC value which provides superior results as compared to existing feature/gene selection methods in terms of multiple performance measures. In the second method, namely the histogram-based clustering and Euclidean distance with fold change-based ranking (HCEDFCR), the miRNAs are first divided into different clusters using a novel histogram-based clustering method. Then, the product of the Euclidean distance and the fold change value, using the control and resistant miRNA expressions, are used to rank the clusters and the miRNAs inside the clusters. Finally, a portion of the miRNAs is selected from the top of the rank list in every cluster. For most of the datasets, the top 20 miR...
Source: Network Modeling Analysis in Health Informatics and Bioinformatics - Category: Bioinformatics Source Type: research