Computational models with thermodynamic and composition features improve siRNA design

BackgroundSmall interfering RNAs (siRNAs) have become an important tool in cell and molecular biology. Reliable design of siRNA molecules is essential for the needs of large functional genomics projects.ResultsTo improve the design of efficient siRNA molecules, we performed a comparative, thermodynamic and correlation analysis on a heterogeneous set of 653 siRNAs collected from the literature. We used this training set to select siRNA features and optimize computational models. We identified 18 parameters that correlate significantly with silencing efficiency. Some of these parameters characterize only the siRNA sequence, while others involve the whole mRNA. Most importantly, we derived an siRNA position-dependent consensus, and optimized the free-energy difference of the 5' and 3' terminal dinucleotides of the siRNA antisense strand. The position-dependent consensus is based on correlation and t-test analyses of the training set, and accounts for both significantly preferred and avoided nucleotides in all sequence positions. On the training set, the two parameters' correlation with silencing efficiency was 0.5 and 0.36, respectively. Among other features, a dinucleotide content index and the frequency of potential targets for siRNA in the mRNA added predictive power to our model (R = 0.55). We showed that our model is effective for predicting the efficiency of siRNAs at different concentrations.We optimized a neural network model on our training set using three parameters characterizing the siRNA sequence, and predicted efficiencies for the test siRNA dataset recently published by Novartis. On this validation set, the correlation coefficient between predicted and observed efficiency was 0.75. Using the same model, we performed a transcriptome-wide analysis of optimal siRNA targets for 22,600 human mRNAs.ConclusionWe demonstrated that the properties of the siRNAs themselves are essential for efficient RNA interference. The 5' ends of antisense strands of efficient siRNAs are U-rich and possess a content similarity to the pyrimidine-rich oligonucleotides interacting with the polypurine RNA tracks that are recognized by RNase H. The advantage of our method over similar methods is the small number of parameters. As a result, our method requires a much smaller training set to produce consistent results. Other mRNA features, though expensive to compute, can slightly improve our model.

[1]  Xiuyuan Hu,et al.  Relative gene-silencing efficiencies of small interfering RNAs targeting sense and antisense transcripts from the same genetic locus. , 2004, Nucleic acids research.

[2]  A D Tsodikov,et al.  Thermodynamic calculations and statistical correlations for oligo-probes design. , 2003, Nucleic acids research.

[3]  A D Tsodikov,et al.  Thermodynamic criteria for high hit rate antisense oligonucleotide design. , 2003, Nucleic acids research.

[4]  C. F. Bennett,et al.  Efficient Reduction of Target RNAs by Small Interfering RNA and RNase H-dependent Antisense Agents , 2003, The Journal of Biological Chemistry.

[5]  Ye Ding,et al.  Sfold web server for statistical folding and rational design of nucleic acids , 2004, Nucleic Acids Res..

[6]  Dieter Huesken,et al.  Design of a genome-wide siRNA library using an artificial neural network , 2005, Nature Biotechnology.

[7]  A. Konagaya,et al.  An Effective Method for Selecting siRNA Target Sequences in Mammalian Cells , 2004, Cell cycle.

[8]  D. Turner,et al.  Incorporating chemical modification constraints into a dynamic programming algorithm for prediction of RNA secondary structure. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[9]  Ola Snøve,et al.  A comparison of siRNA efficacy predictors. , 2004, Biochemical and biophysical research communications.

[10]  T. Du,et al.  Asymmetry in the Assembly of the RNAi Enzyme Complex , 2003, Cell.

[11]  G. Hutvagner,et al.  A microRNA in a Multiple-Turnover RNAi Enzyme Complex , 2002, Science.

[12]  D. Chang,et al.  Using a hydrogen-bond index to predict the gene-silencing efficiency of siRNA based on the local structure of mRNA , 2017, 1710.07413.

[13]  J. Manola,et al.  A library of siRNA duplexes targeting the phosphoinositide 3-kinase pathway: determinants of gene silencing for use in cell-based screens. , 2004, Nucleic acids research.

[14]  J. Krol,et al.  Structural Features of MicroRNA (miRNA) Precursors and Their Relevance to miRNA Biogenesis and Small Interfering RNA/Short Hairpin RNA Design* , 2004, Journal of Biological Chemistry.

[15]  S. Jayasena,et al.  Functional siRNAs and miRNAs Exhibit Strand Bias , 2003, Cell.

[16]  K. Ui-Tei,et al.  Guidelines for the selection of highly effective siRNA sequences for mammalian and chick RNA interference. , 2004, Nucleic acids research.

[17]  Phillip A Sharp,et al.  siRNAs can function as miRNAs , 2003 .

[18]  P. Zamore,et al.  A Protein Sensor for siRNA Asymmetry , 2004, Science.

[19]  Guiliang Tang,et al.  siRNA and miRNA: an insight into RISCs. , 2005, Trends in biochemical sciences.

[20]  D. Barford,et al.  Crystal structure of a PIWI protein suggests mechanisms for siRNA recognition and slicer activity , 2004, The EMBO journal.

[21]  Henning Urlaub,et al.  Single-Stranded Antisense siRNAs Guide Target RNA Cleavage in RNAi , 2002, Cell.

[22]  R. Russell,et al.  Principles of MicroRNA–Target Recognition , 2005, PLoS biology.

[23]  D. Turner,et al.  Improved free-energy parameters for predictions of RNA duplex stability. , 1986, Proceedings of the National Academy of Sciences of the United States of America.

[24]  P. Zamore,et al.  ATP Requirements and Small Interfering RNA Structure in the RNA Interference Pathway , 2001, Cell.

[25]  Lin He,et al.  MicroRNAs: small RNAs with a big role in gene regulation , 2004, Nature Reviews Genetics.

[26]  T. Tuschl,et al.  RNA interference is mediated by 21- and 22-nucleotide RNAs. , 2001, Genes & development.

[27]  E. Lai Micro RNAs are complementary to 3′ UTR sequence motifs that mediate negative post-transcriptional regulation , 2002, Nature Genetics.

[28]  Pål Sætrom,et al.  Predicting the efficacy of short oligonucleotides in antisense and RNAi experiments with boosted genetic programming , 2004, Bioinform..

[29]  T. Tuschl,et al.  Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells , 2001, Nature.

[30]  A. D. Clark,et al.  Crystal structure of HIV‐1 reverse transcriptase in complex with a polypurine tract RNA:DNA , 2001, The EMBO journal.

[31]  Erik L L Sonnhammer,et al.  Improved and automated prediction of effective siRNA. , 2004, Biochemical and biophysical research communications.

[32]  Elisa Izaurralde,et al.  RNAi: finding the elusive endonuclease. , 2004, RNA.

[33]  J. Castle,et al.  Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs , 2005, Nature.

[34]  A. Reynolds,et al.  Rational siRNA design for RNA interference , 2004, Nature Biotechnology.

[35]  Ravi Sachidanandam,et al.  Free energy lights the path toward more effective RNAi , 2003, Nature Genetics.

[36]  D. Turner,et al.  Predicting oligonucleotide affinity to nucleic acid targets. , 1999, RNA.

[37]  Claes Wahlestedt,et al.  A systematic analysis of the silencing effects of an active siRNA at all single-nucleotide mismatched target sites , 2005, Nucleic acids research.

[38]  Georg Sczakiel,et al.  The activity of siRNA in mammalian cells is related to structural target accessibility: a comparison with antisense oligonucleotides. , 2003, Nucleic acids research.

[39]  B. Li,et al.  Expression profiling reveals off-target gene regulation by RNAi , 2003, Nature Biotechnology.

[40]  Tomoyuki Yamada,et al.  siDirect: highly effective, target-specific siRNA design software for mammalian RNA interference , 2004, Nucleic Acids Res..

[41]  Zdenek Moravek,et al.  Efficient RNA interference depends on global context of the target sequence: quantitative analysis of silencing efficiency using Eulerian graph representation of siRNA , 2022 .

[42]  V. Ambros,et al.  The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14 , 1993, Cell.

[43]  D. Turner,et al.  Thermodynamic parameters for an expanded nearest-neighbor model for formation of RNA duplexes with Watson-Crick base pairs. , 1998, Biochemistry.

[44]  J. Sabina,et al.  Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. , 1999, Journal of molecular biology.

[45]  M. Amarzguioui,et al.  An algorithm for selection of functional siRNA sequences. , 2004, Biochemical and biophysical research communications.