are important developmentally, lin-4 and let-7 regulate worm growth. Because the let-7 gene is also found in flies, humans, and other species, the regulatory mechanism of this miRNA is well conserved . Since let-7 is a well-known and typical miRNA, we show its sequence and hairpin structure in Figure 7.4. To identify miRNA genes and targets, computational methods can be developed to perform reliable predictions.
Viral Gene Knockdown RNAi can be used to protect host genome against virus infections by suppressing viral expression. An shRNAwas constructed to silence the niaprotease (Pro) gene of potato virus Y (PVY) in tobacco plants . siRNAs were used successfully to silence the respiratory syncytial virus (RSV), an RNA virus that is responsible for severe respiratory disease in neonates and infants . siRNAs targeting the M-BCR/ABL fusion site were introduced into human K562 cell and specifically silenced leukemic cells . To silence viral genes effectively, we want to silence the target genes maximally and nontarget genes minimally. For these purposes, bioinformatics approaches help optimize siRNA silencing efficacy and specificity.
Gene Function Analysis The function of a gene can be determined by silencing its mRNA and examining the phenotype changes. This loss-of-function method can be used to study gene functions systematically for all genes in an organism, as has been done for C. elegans . RNAi also provides a high-throughput procedure for functional genomics in plants . In addition, loss-of-function of selected genes also helps to build gene regulatory networks and cell signaling pathways . Since loss-of-function studies require maximal target silencing and minimal nontarget silencing, bioinformatics approaches also provide opportunities for optimal solutions with respect to silencing efficacy and specificity.
Disease Treatment RNAi can specifically knock down genes responsible for human diseases, providing a novel means of therapeutics. Chemically modified siRNAs targeting the apoB gene were injected through the tail of mice and reduced their total level of cholesterol . shRNAs were transfected into mice to investigate the therapeutical treatment of dominant polyglutamine expansion diseases, which include Huntington's disease . These experiments in mice demonstrated the therapeutic potential of RNAi for the treatment of human hypercholesterolaemia and Huntington's disease. HIV-1 RNAs were inhibited by siRNAs and shRNAs in human cell lines , showing that HIV virus is amenable to RNAi treatment. In addition to silencing efficacy and specificity, the concentration and temporal changes of the siRNA and the target mRNA should be accessible for efficient transfection. The RNAi kinetics can be modeled by differential equations.
Algorithms Inspired by RNAi Having discussed some biological applications of RNAi, we point out some possibilities of developing algorithms based on RNAi mechanisms. Potentially, the principles of target recognition and degradation of RNAi can be used to design an algorithm for detecting and destroying computer viruses. Once a binary stream of a virus has been identified, it can be used as a virus identifier and fed into a search-deletion engine that detects and eliminates computer codes that are similar to virus codes. Although it has not yet been realized, this idea holds promise for a large open problem in computer science. We anticipate that control of false positivity is a critical issue, since virus codes might share similarity with normal programs. However, it might be a good strategy in certain situations, such as those where all the working programs have known code structures and virus codes are well distinguishable.
In implementing this RNAi-based search-deletion engine, the principle of siRNA amplification of transitive RNAi can also be incorporated. From the code that has been detected and confirmed to be viral, more virus identifiers can be copied from the code itself. These secondary virus identifiers will increase the degradation efficiency. The siRNA amplification mechanism of tRNAi can also be applied to authorship identification and software forensics. Once a piece of work has been determined to be positively related, its content can be used to produce more identifiers to improve search efficiency.
Although RNAi is highly effective, attention must be paid to some particular issues when designing an experiment or conducting a bioinformatic research. RNAi silencing components such as dicer and RISC can themselves be silenced, suppressing the RNAi pathway. For example, potyviruses encode a protein, HC-Pro, that potently disables PTGS by preventing the dsRNA cleavage activity of dicer . Therefore, a plant challenged with a potyvirus becomes a battleground in the fight between a defense and a counterdefense strategy. In addition, although RNAi is a widely conserved mechanism, some organisms, such as Saccharomyces cerevisiae, are RNAi negative due to the lack of RNAi components. Although the metabolism, protein interaction, and cell cycle behaviors of S. cerevisiae have been studied extensively, the lack of RNAi makes it difficult to integrate mRNA regulation into existing networks.
An important issue in RNAi is that when silencing a target gene, it is possible to knock down nontarget genes, causing off-target effects. Although RNAi is generally considered specific, off-target effects exist extensively [27,51]. It is conceivable that silencing a large proportion of organismal genes will cause loss of function of too many genes, making the organism unstable. Therefore, off-target silencing must be carefully controlled. Furthermore, siRNA sequences binding to different regions on a target mRNA have variable silencing capabilities. It is important to predict an siRNA's silencing efficacy, measured via the percent remaining level of the target mRNA, before it is synthesized and used. Rational design rules have been developed to predict this silencing efficacy [4,13,26,29,57,59,66]. To use these rules effectively, siRNA efficacy should be predicted using computational approaches. Related to microRNA biology, miRNA genes and their targets are usually identified computationally to determine their regulatory functions.
These issues are critical for the success of RNAi experiments; they also provide opportunities for researchers in bioinformatics and knowledge discovery. To characterize off-target effects thoroughly, all genes to be silenced, with each gene as target, must be examined. It is too expensive, if feasible, to conduct this total investigation experimentally. Computational simulation can, however, simulate the off-target searches and can extend the parameters beyond experimental conditions.
The binding between siRNA and mRNA allows for mismatches, G-U wobbles, and bulges. But it is not clear how flexible nature allows this binding to be. Simulations using graph models can estimate this critical mismatch. In addition, to better solve the siRNA efficacy problem, machine learning and data mining approaches can be used to improve prediction accuracies. Computational methods not only can predict siRNA efficacy using rational design rules, they can also evaluate the relative significance of different efficacy rules. Furthermore, computational approaches have been developed for the prediction of miRNA genes by predicting possible occurrences of the stem-loop structures. The targets regulated by an miRNA are also predictable by searching genes that possess binding regions based on known miRNA-target complementarity. In the following sections we discuss these issues and their computational solutions in detail.
Was this article helpful?
Discover secrets, myths, truths, lies and strategies for dealing effectively with cholesterol, now and forever! Uncover techniques, remedies and alternative for lowering your cholesterol quickly and significantly in just ONE MONTH! Find insights into the screenings, meanings and numbers involved in lowering cholesterol and the implications, consideration it has for your lifestyle and future!