Silencer Select design algorithm improves effective siRNA prediction accuracy
Current siRNA design algorithms predict effective siRNAs that induce 70% target mRNA knockdown with only ~80% confidence and are inadequate for predicting more efficient siRNAs. Many RNAi applications demand better efficiency than current algorithms offer. Therefore, we used a powerful machine learning method and performance data from thousands of siRNAs to better understand the link between an siRNA’s sequence, target location, and thermodynamic properties and its silencing efficiency. The result is the Silencer Select siRNA Design Algorithm.
Silencer Select siRNA Design Algorithm Significantly Improves Effective siRNA Prediction Accuracy.
The Silencer Select siRNA design algorithm was used to design 155 siRNAs to 40 different targets. These siRNAs were tested side by side with siRNAs designed using the previous algorithm at 5 nM in HeLa cells. mRNA knockdown was measured 48 h post-transfection via qRT-PC R using TaqMan Gene Expression Assays. Results are expressed as percent of mRNA remaining compared to Silencer Negative Control #1 siRNA treated cells. The inset shows the percentage of siRNAs that elicited ≥70% and ≥80% mRNA knockdown.
New Silencer Select Design Algorithm
- Incorporates >90 different sequence and thermodynamic parameters
- Increases predictive accuracy 28% over previous generation siRNA design
- Yields siRNAs that are up to 100 fold more potent than both modified and
- unmodified siRNAs from other suppliers
- Provides significantly higher percentage of “on-target” phenotypes compared to other siRNAs