Soft Computing Approach for Optimization of siRNA Efficiency Prediction for Post transcriptional Gene Silencing

Abstract

Post-transcriptional gene silencing by RNA interference is newlinemediated by small interfering RNA called siRNA. This gene newlinesilencing mechanism can be exploited therapeutically to a wide newlinevariety of disease-associated targets, especially in AIDS, newlineneurodegenerative diseases, cholesterol and cancer on mice with the newlinehope of extending these approaches to treat humans. Over the recent newlinepast, a significant amount of work has been undertaken to understand newlinethe gene silencing mediated by exogenous siRNA. The design of newlineefficient exogenous siRNA sequences is challenging because of many newlineissues related to siRNA. While designing efficient siRNA, target newlinemRNAs must be selected such that their corresponding siRNAs are newlinelikely to be efficient against that target and unlikely to accidentally newlinesilence other transcripts due to sequence similarity. So before doing newlinegene silencing by siRNAs, it is essential to analyze their off-target newlineeffects in addition to their inhibition efficiency against a particular newlinetarget. Hence designing exogenous siRNA with good knock-down newlineefficiency and target specificity is an area of concern to be addressed. newlineSome methods have been developed already by considering both newlineinhibition efficiency and off-target possibility of siRNA against a gene. Out of these methods, only a few have achieved good inhibition newlineefficiency, specificity and sensitivity.The main focus of this thesis is to develop computational newlinemethods to optimize the efficiency of siRNA in terms of inhibition newlinecapacity and off-target possibility against target mRNAs with newlineimproved efficacy, which may be useful in the area of gene silencing newlineand drug design for tumor development. This study aims to newlineinvestigate the currently available siRNA prediction approaches and newlineto devise a better computational approach to tackle the problem of newlinesiRNA efficacy by inhibition capacity and off-target possibility. The newlinestrength and limitations of the available approaches are investigated newlineand taken into consideration for making improved solution. Thus the newlineapproaches proposed in this stumodel, named siRNA Designer, is used for optimizing the inhibition newlineefficiency of siRNA against target genes. The second ANN model, newlinenamed Optimized siRNA Designer, OpsiD, produces efficient newlinesiRNAs with high inhibition efficiency to degrade target genes with newlineimproved sensitivity-specificity, and identifies the off-target knockdown newlinepossibility of siRNA against non-target genes. The models are newlinetrained and tested against a large data set of siRNA sequences. The newlinevalidations are conducted using Pearson Correlation Coefficient, newlineMathews Correlation Coefficient, Receiver Operating Characteristic newlineanalysis, Accuracy of prediction, Sensitivity and Specificity.