Why customized siDesigner?
Customized siDesigner is a siRNA designing tool that has been customized to design siRNAs with higher potency and specificity. We evaluated a number of well established siRNA designing parameters on a set of biologically validated siRNAs to examine their efficiency in identifying highly functional siRNAs from non-functional ones and used SVM classifier to classify between efficient and non-efficient siRNAs based on a parameter set that showed better performance on the above study. To improve target specificity, apart from restricting the number of potential off-targets (which is the approach taken by present day siRNA designing tools) we applied a novel approach to filter out functionally similar or related off-targets based on the Gene Ontology process terms of the direct target and each of the predicted off-targets. Moreover we give user a choice to interactively select one or more off-targets that he thinks can be more harmful, to filter out- so that it shall not be targeted any more. This tool is first in its kind to provide the user with such choice for better siRNA target specificity.
What is functional off-target filtering and why it is important?
Functional off-target filtering is a unique approach of Customized siDesigner to predict off-targets that have functional relation or similarity with the direct target and avoid them from being off-targeted. This approach is new to be considered while designing siRNAs but very important because if a selected siRNA inadvertently targets some unintended transcript that is functionally similar or related to the biochemical pathway of the direct target gene, that may lead to misleading false positive phenotype and ruin the experimental findings.
How the siRNAs are selected for targeting a specified gene?
The siRNAs are selected using a set of well established siRNA designing parameters- along with considerations for homology filtering, common target for all alternatively spliced isoforms for a given gene and microRNA like off-target filtering. siRNAs can be targetd to the coding region or the 3’ UTR of an mRNA. A support vector machine is used to classify potent siRNAs targeting the specified gene using eight siRNA selection parameters. This is followed by the homology check, microRNA like off-targets check and functional off-target filtering. Finally a score is generated for the selected siRNAs based on their potency and specificity. Top 30 siRNAs according to this score is then displayed to the user.
How can I control the stringency of the siRNA selection parameters?
The stringency of the siRNA selection parameters can be controlled by user as the user interface provides choices to tick or omit specific parameters to be strictly considered. User can control parameters like low GC% content, weaker 5’ end thermodynamic energy of antisense strand, absence of 4 or more identical nucleotides in a row, homology filter, SNP filter, immune stimulatory motif filter, microRNA like off-target filter and functional off-target filter.
Why use SVM for classification of efficient siRNAs?
SVM is an excellent classifier for complex data and has been used broadly in biological studies. Trained with adequate number of biologically validated data, the SVM classifier performs very well when it comes to classifying new examples. In our case, the performance of the SVM classifier was consistently good for four different datasets from four different sources.
How can I predict the efficiency of the designed siRNAs?
The final selected siRNAs are given a score to reflect both their potency and specificity.
How to use Customized siDesigner?
Customized siDesigner takes the target gene symbol as input and outputs the topmost potent siRNAs for the target gene. The target region within mRNA and stringent siRNA selection parameters can be selected.
How can I use the feedback process for filtering out some specific off-targets?
After the selected siRNAs are displayed, user can select one siRNA to view its predicted off-target list. He can select one or more off-targets from the list to be filtered out. The program runs again to produce another list of siRNAs that do not have the selected off-targets.
What is the advantage of the user feedback process?
The user feedback process enables user to choose one or more off-targets to be filtered out so that they cannot be targeted by the selected siRNAs. The selected siRNAs may inadvertently target some transcripts that are related to the function or biochemical processes in which the direct target is involved. Or they can be involved in cellular metabolic processes. Such off-targeting may lead to misleading phenotypes or very harmful consequences. To avoid this we give user a choice to carefully avoid such potential harmful off-targets by an interactive feedback process.
I do not get any output siRNAs because the target gene symbol cannot be found. What is wrong?
We used current Refseq annotations for human mRNA transcripts. If a gene symbol cannot be found then please check in NCBI for its current symbol and annotation. For further problems please contact us at firstname.lastname@example.org.
I do not get any output siRNAs in spite of giving the correct gene symbol. What should I do?
If no siRNAs are selected then you may have to loosen the stringency of the constraints. Please deselect a couple of stringent parameters and try to run the program again. For further problems please contact email@example.com.
How Customized siDesigner does take care of alternatively spliced isoforms for the target gene?
Customized siDesigner is designed to target every mRNA transcripts for a given gene. Presently we do not consider siRNA designing for one or more specific transcripts. We wish to incorporate this feature in future versions.
Please cite Customized siDesigner:
Das S, Ghosal S, Kozak K, Chakrabarti J. (2012) An siRNA designing tool with a unique functional off-target filtering approach. J Biomol Struct Dyn. DOI: 10.1080/07391102.2012.736758