Improving model predictions for RNA interference activities that use support vector machine regression by combining and filtering features

Protection of animals against devastating RNA viruses using CRISPR-Cas13s

The intrinsic nature of CRISPR-Cas in conferring immunity to bacteria and archaea has been repurposed to combat pathogenic agents in mammalian and plant cells. In this regard, CRISPR-Cas13 systems have proved their remarkable potential for single-strand RNA viruses targeting. Here, different types of Cas13 orthologs were applied to knockdown foot-and-mouth disease virus (FMDV), a highly contagious disease of a wide variety of species with genetically diverse strains and is widely geographically distributed. Using programmable CRISPR RNAs capable of targeting conserved regions of the viral genome, all Cas13s from CRISPR system type VI (subtype A/B/D) could comprehensively target and repress different serotypes of FMDV virus. This approach has the potential to destroy all strains of a virus as targets the ultra-conserved regions of genome. We experimentally compared the silencing efficiency of CRISPR and RNAi by designing the most effective short hairpin RNAs according to our developed scoring system and observed comparable results. This study showed successful usage of various Cas13 enzymes for suppression of FMDV, which provides a flexible strategy to battle with other animal infectious RNA viruses, an underdeveloped field in the biotechnology scope.

Constructing the boundary between potent and ineffective siRNAs by MG-algorithm with C-features

BACKGROUND

In siRNA based antiviral therapeutics, selection of potent siRNAs is an indispensable step, but these commonly used features are unable to construct the boundary between potent and ineffective siRNAs.

RESULTS

Here, we select potent siRNAs by removing ineffective ones, where these conditions for removals are constructed by C-features of siRNAs, C-features are generated by MG-algorithm, Icc-cluster and the different combinations of some commonly used features, MG-algorithm and Icc-cluster are two different algorithms to search the nearest siRNA neighbors. For the ineffective siRNAs in test data, they are removed from test data by I-iteration, where I-iteration continually updates training data by adding these successively removed siRNAs. Furthermore, the efficacy of siRNAs of test data is predicted by their nearest neighbors of training data.

CONCLUSIONS

By siRNAs of Hencken dataset, results show that our algorithm removes almost ineffective siRNAs from test data, gives the clear boundary between potent and ineffective siRNAs, and accurately predicts the efficacy of siRNAs also. We suggest that our algorithm can provide new insights for selecting the potent siRNAs.

siRNA Design and GalNAc-Empowered Hepatic Targeted Delivery

Small interfering RNA (siRNA) is a clinically approved therapeutic modality, which has attracted widespread attention not only from basic research but also from pharmaceutical industry. As siRNA can theoretically modulate any disease-related gene's expression, plenty of siRNA therapeutic pipelines have been established by tens of biotechnology companies. The drug performance of siRNA heavily depends on the sequence, the chemical modification, and the delivery of siRNA. Here, we describe the rational design protocol of siRNA, and provide some modification patterns that can enhance siRNA's stability and reduce its off-target effect. Also, the delivery method based on N-acetylgalactosamine (GalNAc)-siRNA conjugate that is widely employed to develop therapeutic regimens for liver-related diseases is also recapitulated.

Using the Chou’s Pseudo Component to Predict the ncRNA Locations Based on the Improved K-Nearest Neighbor (iKNN) Classifier

Background:

The non-coding RNA identification at the organelle genome level is a challenging task. In our previous work, an ncRNA dataset with less than 80% sequence identity was built, and a method incorporating an increment of diversity combining with support vector machine method was proposed.

Objective:

Based on the ncRNA_361 dataset, a novel decision-making method-an improved KNN (iKNN) classifier was proposed.

Methods:

In this paper, based on the iKNN algorithm, the physicochemical features of nucleotides, the degeneracy of genetic codons, and topological secondary structure were selected to represent the effective ncRNA characters. Then, the incremental feature selection method was utilized to optimize the feature set.

Results:

The results of iKNN indicated that the decision-making method of mean value is distinctly superior to the traditional decision-making method of majority vote the Increment of Diversity Combining Support Vector Machine (ID-SVM). The iKNN algorithm achieved an overall accuracy of 97.368% in the jackknife test, when k=3.

Conclusion:

It should be noted that the triplets of the structure-sequence mode under reading frames not only contains the entire sequence information but also reflects whether the base was paired or not, and the secondary structural topological parameters further describe the ncRNA secondary structure on the spatial level. The ncRNA dataset and the iKNN classifier are freely available at http://202.207.14.87:8032/fuwu/iKNN/index.asp.

Integrating Artificial Intelligence and Nanotechnology for Precision Cancer Medicine

Artificial intelligence (AI) and nanotechnology are two fields that are instrumental in realizing the goal of precision medicine-tailoring the best treatment for each cancer patient. Recent conversion between these two fields is enabling better patient data acquisition and improved design of nanomaterials for precision cancer medicine. Diagnostic nanomaterials are used to assemble a patient-specific disease profile, which is then leveraged, through a set of therapeutic nanotechnologies, to improve the treatment outcome. However, high intratumor and interpatient heterogeneities make the rational design of diagnostic and therapeutic platforms, and analysis of their output, extremely difficult. Integration of AI approaches can bridge this gap, using pattern analysis and classification algorithms for improved diagnostic and therapeutic accuracy. Nanomedicine design also benefits from the application of AI, by optimizing material properties according to predicted interactions with the target drug, biological fluids, immune system, vasculature, and cell membranes, all affecting therapeutic efficacy. Here, fundamental concepts in AI are described and the contributions and promise of nanotechnology coupled with AI to the future of precision cancer medicine are reviewed.

A review on current status of antiviral siRNA

Viral diseases like influenza, AIDS, hepatitis, and Ebola cause severe epidemics worldwide. Along with their resistant strains, new pathogenic viruses continue to be discovered so creating an ongoing need for new antiviral treatments. RNA interference is a cellular gene‐silencing phenomenon in which sequence‐specific degradation of target mRNA is achieved by means of complementary short interfering RNA (siRNA) molecules. Short interfering RNA technology affords a potential tractable strategy to combat viral pathogenesis because siRNAs are specific, easy to design, and can be directed against multiple strains of a virus by targeting their conserved gene regions. In this review, we briefly summarize the current status of siRNA therapy for representative examples from different virus families. In addition, other aspects like their design, delivery, medical significance, bioinformatics resources, and limitations are also discussed.

PClass: Protein Quaternary Structure Classification by Using Bootstrapping Strategy as Model Selection

Protein quaternary structure complex is also known as a multimer, which plays an important role in a cell. The dimer structure of transcription factors is involved in gene regulation, but the trimer structure of virus-infection-associated glycoproteins is related to the human immunodeficiency virus. The classification of the protein quaternary structure complex for the post-genome era of proteomics research will be of great help. Classification systems among protein quaternary structures have not been widely developed. Therefore, we designed the architecture of a two-layer machine learning technique in this study, and developed the classification system PClass. The protein quaternary structure of the complex is divided into five categories, namely, monomer, dimer, trimer, tetramer, and other subunit classes. In the framework of the bootstrap method with a support vector machine, we propose a new model selection method. Each type of complex is classified based on sequences, entropy, and accessible surface area, thereby generating a plurality of feature modules. Subsequently, the optimal model of effectiveness is selected as each kind of complex feature module. In this stage, the optimal performance can reach as high as 70% of Matthews correlation coefficient (MCC). The second layer of construction combines the first-layer module to integrate mechanisms and the use of six machine learning methods to improve the prediction performance. This system can be improved over 10% in MCC. Finally, we analyzed the performance of our classification system using transcription factors in dimer structure and virus-infection-associated glycoprotein in trimer structure. PClass is available via a web interface at http://predictor.nchu.edu.tw/PClass/.

ASPsiRNA: A Resource of ASP-siRNAs Having Therapeutic Potential for Human Genetic Disorders and Algorithm for Prediction of Their Inhibitory Efficacy

Allele-specific siRNAs (ASP-siRNAs) have emerged as promising therapeutic molecules owing to their selectivity to inhibit the mutant allele or associated single-nucleotide polymorphisms (SNPs) sparing the expression of the wild-type counterpart. Thus, a dedicated bioinformatics platform encompassing updated ASP-siRNAs and an algorithm for the prediction of their inhibitory efficacy will be helpful in tackling currently intractable genetic disorders. In the present study, we have developed the ASPsiRNA resource (http://crdd.osdd.net/servers/aspsirna/) covering three components viz (i) ASPsiDb, (ii) ASPsiPred, and (iii) analysis tools like ASP-siOffTar. ASPsiDb is a manually curated database harboring 4543 (including 422 chemically modified) ASP-siRNAs targeting 78 unique genes involved in 51 different diseases. It furnishes comprehensive information from experimental studies on ASP-siRNAs along with multidimensional genetic and clinical information for numerous mutations. ASPsiPred is a two-layered algorithm to predict efficacy of ASP-siRNAs for fully complementary mutant (Eff^mut) and wild-type allele (Eff^wild) with one mismatch by ASPsiPred^SVM and ASPsiPred^matrix, respectively. In ASPsiPred^SVM, 922 unique ASP-siRNAs with experimentally validated quantitative Eff^mut were used. During 10-fold cross-validation (10nCV) employing various sequence features on the training/testing dataset (T737), the best predictive model achieved a maximum Pearson’s correlation coefficient (PCC) of 0.71. Further, the accuracy of the classifier to predict Eff^mut against novel genes was assessed by leave one target out cross-validation approach (LOTOCV). ASPsiPred^matrix was constructed from rule-based studies describing the effect of single siRNA:mRNA mismatches on the efficacy at 19 different locations of siRNA. Thus, ASPsiRNA encompasses the first database, prediction algorithm, and off-target analysis tool that is expected to accelerate research in the field of RNAi-based therapeutics for human genetic diseases.

AVP-IC50 Pred: Multiple machine learning techniques-based prediction of peptide antiviral activity in terms of half maximal inhibitory concentration (IC50)

Peptide-based antiviral therapeutics has gradually paved their way into mainstream drug discovery research. Experimental determination of peptides' antiviral activity as expressed by their IC50 values involves a lot of effort. Therefore, we have developed "AVP-IC50 Pred," a regression-based algorithm to predict the antiviral activity in terms of IC50 values (μM). A total of 759 non-redundant peptides from AVPdb and HIPdb were divided into a training/test set having 683 peptides (T(683)) and a validation set with 76 independent peptides (V(76)) for evaluation. We utilized important peptide sequence features like amino-acid compositions, binary profile of N8-C8 residues, physicochemical properties and their hybrids. Four different machine learning techniques (MLTs) namely Support vector machine, Random Forest, Instance-based classifier, and K-Star were employed. During 10-fold cross validation, we achieved maximum Pearson correlation coefficients (PCCs) of 0.66, 0.64, 0.56, 0.55, respectively, for the above MLTs using the best combination of feature sets. All the predictive models also performed well on the independent validation dataset and achieved maximum PCCs of 0.74, 0.68, 0.59, 0.57, respectively, on the best combination of feature sets. The AVP-IC50 Pred web server is anticipated to assist the researchers working on antiviral therapeutics by enabling them to computationally screen many compounds and focus experimental validation on the most promising set of peptides, thus reducing cost and time efforts. The server is available at http://crdd.osdd.net/servers/ic50avp.

Computational design of artificial RNA molecules for gene regulation

RNA interference (RNAi) is a powerful tool for the regulation of gene expression. Small exogenous noncoding RNAs (ncRNAs) such as siRNA and shRNA are the active silencing agents, intended to target and cleave complementary mRNAs in a specific way. They are widely and successfully employed in functional studies, and several ongoing and already completed siRNA-based clinical trials suggest encouraging results in the regulation of overexpressed genes in disease. siRNAs share many aspects of their biogenesis and function with miRNAs, small ncRNA molecules transcribed from endogenous genes which are able to repress the expression of target mRNAs by either inhibiting their translation or promoting their degradation. Although siRNA and artificial miRNA molecules can significantly reduce the expression of overexpressed target genes, cancer and other diseases can also be triggered or sustained by upregulated miRNAs. Thus, in the past recent years, molecular tools for miRNA silencing, such as antagomiRs and miRNA sponges, have been developed. These molecules have shown their efficacy in the derepression of genes downregulated by overexpressed miRNAs. In particular, while a single antagomiR is able to inhibit a single complementary miRNA, an artificial sponge construct usually contains one or more binding sites for one or more miRNAs and functions by competing with the natural targets of these miRNAs. As a consequence, natural miRNA targets are reexpressed at their physiological level. In this chapter we review the most successful methods for the computational design of siRNAs, antagomiRs, and miRNA sponges and describe the most popular tools that implement them.

VIRsiRNApred: a web server for predicting inhibition efficacy of siRNAs targeting human viruses

BACKGROUND

Selection of effective viral siRNA is an indispensable step in the development of siRNA based antiviral therapeutics. Despite immense potential, a viral siRNA efficacy prediction algorithm is still not available. Moreover, performances of the existing general mammalian siRNA efficacy predictors are not satisfactory for viral siRNAs. Therefore, we have developed "VIRsiRNApred" a support vector machine (SVM) based method for predicting the efficacy of viral siRNA.

METHODS

In the present study, we have employed a new dataset of 1725 viral siRNAs with experimentally verified quantitative efficacies tested under heterogeneous experimental conditions and targeting as many as 37 important human viruses including HIV, Influenza, HCV, HBV, SARS etc. These siRNAs were divided into training (T1380) and validation (V345) datasets. Important siRNA sequence features including mono to penta nucleotide frequencies, binary pattern, thermodynamic properties and secondary structure were employed for model development.

RESULTS

During 10-fold cross validation on T1380 using hybrid approach, we achieved a maximum Pearson Correlation Coefficient (PCC) of 0.55 between predicted and actual efficacy of viral siRNAs. On V345 independent dataset, our best model achieved a maximum correlation of 0.50 while existing general siRNA prediction methods showed PCC from 0.05 to 0.18. However, using leave one out cross validation PCC was improved to 0.58 and 0.55 on training and validation datasets respectively. SVM performed better than other machine learning techniques used like ANN, KNN and REP Tree.

CONCLUSION

VIRsiRNApred is the first algorithm for predicting inhibition efficacy of viral siRNAs which is developed using experimentally verified viral siRNAs. We hope this algorithm would be useful in predicting highly potent viral siRNA to aid siRNA based antiviral therapeutics development. The web server is freely available at http://crdd.osdd.net/servers/virsirnapred/.

The effect of regions flanking target site on siRNA potency

For a successful RNA interference (RNAi) experiment, selecting the small interference RNA (siRNA) candidates which maximize the knock down effect of the given gene is the critical step. Although various computational approaches have been attempted, the design of efficient siRNA candidates is far from satisfactory yet. In this study, we proposed a novel feature selection algorithm of combined random forest and support vector machine to predict active siRNAs. Using a publically available dataset, we demonstrated that the predictive accuracy would be markedly improved when the context sequence features outside the target site were included. The Pearson correlation coefficient for regression is as high as 0.721, compared to 0.671, 0.668, 0.680, and 0.645, for Biopredsi, i-score, ThermoComposition21 and DSIR, respectively. It revealed that siRNA-target interaction requires appropriate sequence context not only in the target site but also in a broad region flanking the target site.

Greedy feature selection for glycan chromatography data with the generalized Dirichlet distribution

Background

Glycoproteins are involved in a diverse range of biochemical and biological processes. Changes in protein glycosylation are believed to occur in many diseases, particularly during cancer initiation and progression. The identification of biomarkers for human disease states is becoming increasingly important, as early detection is key to improving survival and recovery rates. To this end, the serum glycome has been proposed as a potential source of biomarkers for different types of cancers.

High-throughput hydrophilic interaction liquid chromatography (HILIC) technology for glycan analysis allows for the detailed quantification of the glycan content in human serum. However, the experimental data from this analysis is compositional by nature. Compositional data are subject to a constant-sum constraint, which restricts the sample space to a simplex. Statistical analysis of glycan chromatography datasets should account for their unusual mathematical properties.

As the volume of glycan HILIC data being produced increases, there is a considerable need for a framework to support appropriate statistical analysis. Proposed here is a methodology for feature selection in compositional data. The principal objective is to provide a template for the analysis of glycan chromatography data that may be used to identify potential glycan biomarkers.

Results

A greedy search algorithm, based on the generalized Dirichlet distribution, is carried out over the feature space to search for the set of “grouping variables” that best discriminate between known group structures in the data, modelling the compositional variables using beta distributions. The algorithm is applied to two glycan chromatography datasets. Statistical classification methods are used to test the ability of the selected features to differentiate between known groups in the data. Two well-known methods are used for comparison: correlation-based feature selection (CFS) and recursive partitioning (rpart). CFS is a feature selection method, while recursive partitioning is a learning tree algorithm that has been used for feature selection in the past.

Conclusions

The proposed feature selection method performs well for both glycan chromatography datasets. It is computationally slower, but results in a lower misclassification rate and a higher sensitivity rate than both correlation-based feature selection and the classification tree method.

What parameters to consider and which software tools to use for target selection and molecular design of small interfering RNAs

The design of small gene silencing RNAs with a high probability of being efficient still has some elements of an art, especially when the lowest concentration of small molecules needs to be utilized. The design of highly target-specific small interfering RNAs or short hairpin RNAs is even a greater challenging task. Some logical schemes and software tools that can be used for simplifying both tasks are presented here. In addition, sequence motifs and sequence composition biases of small interfering RNAs that have to be avoided because of specificity concerns are also detailed.

Reconsideration of in silico siRNA design from a perspective of heterogeneous data integration: problems and solutions

The success of RNA interference (RNAi) depends on the interaction between short interference RNAs (siRNAs) and mRNAs. Design of highly efficient and specific siRNAs has become a challenging issue in applications of RNAi. Here, we present a detailed survey on the state-of-the-art siRNAs design, focusing on several key issues with the current in silico RNAi studies, including: (i) inconsistencies among the proposed guidelines for siRNAs design and the incomplete list of siRNAs features, (ii) improper integration of the heterogeneous cross-platform siRNAs data, (iii) inadequate consideration of the binding specificity of the target mRNAs and (iv) reduction in the 'off-target' effect in siRNAs design. With these considerations, the popular in silico siRNAs design rules are reexamined and several inconsistent viewpoints toward siRNAs feature identifications are clarified. In addition, novel computational models for siRNAs design using state-of-art machine learning techniques are discussed, which focus on heterogeneous data integration, joint feature selection and customized siRNAs screening toward highly specific targets. We believe that addressing such issues in siRNA study will provide new clues for further improved design of more efficient and specific siRNAs in RNAi.

Reconsideration of in-silico siRNA design based on feature selection: a cross-platform data integration perspective

RNA interference via exogenous short interference RNAs (siRNA) is increasingly more widely employed as a tool in gene function studies, drug target discovery and disease treatment. Currently there is a strong need for rational siRNA design to achieve more reliable and specific gene silencing; and to keep up with the increasing needs for a wider range of applications. While progress has been made in the ability to design siRNAs with specific targets, we are clearly at an infancy stage towards achieving rational design of siRNAs with high efficacy. Among the many obstacles to overcome, lack of general understanding of what sequence features of siRNAs may affect their silencing efficacy and of large-scale homogeneous data needed to carry out such association analyses represents two challenges. To address these issues, we investigated a feature-selection based in-silico siRNA design from a novel cross-platform data integration perspective. An integration analysis of 4,482 siRNAs from ten meta-datasets was conducted for ranking siRNA features, according to their possible importance to the silencing efficacy of siRNAs across heterogeneous data sources. Our ranking analysis revealed for the first time the most relevant features based on cross-platform experiments, which compares favorably with the traditional in-silico siRNA feature screening based on the small samples of individual platform data. We believe that our feature ranking analysis can offer more creditable suggestions to help improving the design of siRNA with specific silencing targets. Data and scripts are available at http://csbl.bmb.uga.edu/publications/materials/qiliu/siRNA.html.

Progress toward in vivo use of siRNAs-II

RNA interference (RNAi) has been extensively employed for in vivo research since its use was first demonstrated in mammalian cells 10 years ago. Design rules have improved, and it is now routinely possible to obtain reagents that suppress expression of any gene desired. At the same time, increased understanding of the molecular basis of unwanted side effects has led to the development of chemical modification strategies that mitigate these concerns. Delivery remains the single greatest hurdle to widespread adoption of in vivo RNAi methods. However, exciting advances have been made and new delivery systems under development may help to overcome these barriers. This review discusses advances in RNAi biochemistry and biology that impact in vivo use and provides an overview of select publications that demonstrate interesting applications of these principles. Emphasis is placed on work with synthetic, small interfering RNAs (siRNAs) published since the first installment of this review which appeared in 2006.

siPRED: predicting siRNA efficacy using various characteristic methods

Small interfering RNA (siRNA) has been used widely to induce gene silencing in cells. To predict the efficacy of an siRNA with respect to inhibition of its target mRNA, we developed a two layer system, siPRED, which is based on various characteristic methods in the first layer and fusion mechanisms in the second layer. Characteristic methods were constructed by support vector regression from three categories of characteristics, namely sequence, features, and rules. Fusion mechanisms considered combinations of characteristic methods in different categories and were implemented by support vector regression and neural networks to yield integrated methods. In siPRED, the prediction of siRNA efficacy through integrated methods was better than through any method that utilized only a single method. Moreover, the weighting of each characteristic method in the context of integrated methods was established by genetic algorithms so that the effect of each characteristic method could be revealed. Using a validation dataset, siPRED performed better than other predictive systems that used the scoring method, neural networks, or linear regression. Finally, siPRED can be improved to achieve a correlation coefficient of 0.777 when the threshold of the whole stacking energy is ≥-34.6 kcal/mol. siPRED is freely available on the web at http://predictor.nchu.edu.tw/siPRED.

Informatics approach to the rational design of siRNA libraries

This chapter surveys the literature for state-of-the-art methods for the rational design of siRNA libraries. It identifies and presents major milestones in the field of computational modeling of siRNA's gene silencing efficacy. Commonly used features of siRNAs are summarized along with major machine learning techniques employed to build the predictive models. It has also outlined several web-enabled siRNA design tools. To face the challenge of modeling and rational design of chemically modified siRNAs, it also proposes a new cheminformatics approach for the representation and characterization of siRNA molecules. Some preliminary results with this new approach are presented to demonstrate the promising potential of this method for the modeling of siRNA's efficacy. Together with novel delivery technologies and chemical modification techniques, rational siRNA design algorithms will ultimately contribute to chemical biology research and the efficient development of siRNA therapeutics.

Prediction of siRNA potency using sparse logistic regression

RNA interference (RNAi) can modulate gene expression at post-transcriptional as well as transcriptional levels. Short interfering RNA (siRNA) serves as a trigger for the RNAi gene inhibition mechanism, and therefore is a crucial intermediate step in RNAi. There have been extensive studies to identify the sequence characteristics of potent siRNAs. One such study built a linear model using LASSO (Least Absolute Shrinkage and Selection Operator) to measure the contribution of each siRNA sequence feature. This model is simple and interpretable, but it requires a large number of nonzero weights. We have introduced a novel technique, sparse logistic regression, to build a linear model using single-position specific nucleotide compositions which has the same prediction accuracy of the linear model based on LASSO. The weights in our new model share the same general trend as those in the previous model, but have only 25 nonzero weights out of a total 84 weights, a 54% reduction compared to the previous model. Contrary to the linear model based on LASSO, our model suggests that only a few positions are influential on the efficacy of the siRNA, which are the 5' and 3' ends and the seed region of siRNA sequences. We also employed sparse logistic regression to build a linear model using dual-position specific nucleotide compositions, a task LASSO is not able to accomplish well due to its high dimensional nature. Our results demonstrate the superiority of sparse logistic regression as a technique for both feature selection and regression over LASSO in the context of siRNA design.

Multi-task learning for cross-platform siRNA efficacy prediction: an in-silico study

BACKGROUND

Gene silencing using exogenous small interfering RNAs (siRNAs) is now a widespread molecular tool for gene functional study and new-drug target identification. The key mechanism in this technique is to design efficient siRNAs that incorporated into the RNA-induced silencing complexes (RISC) to bind and interact with the mRNA targets to repress their translations to proteins. Although considerable progress has been made in the computational analysis of siRNA binding efficacy, few joint analysis of different RNAi experiments conducted under different experimental scenarios has been done in research so far, while the joint analysis is an important issue in cross-platform siRNA efficacy prediction. A collective analysis of RNAi mechanisms for different datasets and experimental conditions can often provide new clues on the design of potent siRNAs.

RESULTS

An elegant multi-task learning paradigm for cross-platform siRNA efficacy prediction is proposed. Experimental studies were performed on a large dataset of siRNA sequences which encompass several RNAi experiments recently conducted by different research groups. By using our multi-task learning method, the synergy among different experiments is exploited and an efficient multi-task predictor for siRNA efficacy prediction is obtained. The 19 most popular biological features for siRNA according to their jointly importance in multi-task learning were ranked. Furthermore, the hypothesis is validated out that the siRNA binding efficacy on different messenger RNAs(mRNAs) have different conditional distribution, thus the multi-task learning can be conducted by viewing tasks at an "mRNA"-level rather than at the "experiment"-level. Such distribution diversity derived from siRNAs bound to different mRNAs help indicate that the properties of target mRNA have important implications on the siRNA binding efficacy.

CONCLUSIONS

The knowledge gained from our study provides useful insights on how to analyze various cross-platform RNAi data for uncovering of their complex mechanism.

Selection of hyperfunctional siRNAs with improved potency and specificity

One critical step in RNA interference (RNAi) experiments is to design small interfering RNAs (siRNAs) that can greatly reduce the expression of the target transcripts, but not of other unintended targets. Although various statistical and computational approaches have been attempted, this remains a challenge facing RNAi researchers. Here, we present a new experimentally validated method for siRNA design. By analyzing public siRNA data and focusing on hyperfunctional siRNAs, we identified a set of sequence features as potency selection criteria to build an siRNA design algorithm with support vector machines. Additional bioinformatics filters were also included in the algorithm to increase RNAi specificity by reducing potential sequence cross-hybridization or microRNA-like effects. Independent validation experiments were performed, which indicated that the newly designed siRNAs have significantly improved performance, and worked effectively even at low concentrations. Furthermore, our cell-based studies demonstrated that the siRNA off-target effects were significantly reduced when the siRNAs were delivered into cells at the 3 nM concentration compared to 30 nM. Thus, the capability of our new design program to select highly potent siRNAs also renders increased RNAi specificity because these siRNAs can be used at a much lower concentration. The siRNA design web server is available at http://www5.appliedbiosystems.com/tools/siDesign/.

Comparing artificial neural networks, general linear models and support vector machines in building predictive models for small interfering RNAs

Background

Exogenous short interfering RNAs (siRNAs) induce a gene knockdown effect in cells by interacting with naturally occurring RNA processing machinery. However not all siRNAs induce this effect equally. Several heterogeneous kinds of machine learning techniques and feature sets have been applied to modeling siRNAs and their abilities to induce knockdown. There is some growing agreement to which techniques produce maximally predictive models and yet there is little consensus for methods to compare among predictive models. Also, there are few comparative studies that address what the effect of choosing learning technique, feature set or cross validation approach has on finding and discriminating among predictive models.

Principal Findings

Three learning techniques were used to develop predictive models for effective siRNA sequences including Artificial Neural Networks (ANNs), General Linear Models (GLMs) and Support Vector Machines (SVMs). Five feature mapping methods were also used to generate models of siRNA activities. The 2 factors of learning technique and feature mapping were evaluated by complete 3×5 factorial ANOVA. Overall, both learning techniques and feature mapping contributed significantly to the observed variance in predictive models, but to differing degrees for precision and accuracy as well as across different kinds and levels of model cross-validation.

Conclusions

The methods presented here provide a robust statistical framework to compare among models developed under distinct learning techniques and feature sets for siRNAs. Further comparisons among current or future modeling approaches should apply these or other suitable statistically equivalent methods to critically evaluate the performance of proposed models. ANN and GLM techniques tend to be more sensitive to the inclusion of noisy features, but the SVM technique is more robust under large numbers of features for measures of model precision and accuracy. Features found to result in maximally predictive models are not consistent across learning techniques, suggesting care should be taken in the interpretation of feature relevance. In the models developed here, there are statistically differentiable combinations of learning techniques and feature mapping methods where the SVM technique under a specific combination of features significantly outperforms all the best combinations of features within the ANN and GLM techniques.

Approximate Bayesian feature selection on a large meta-dataset offers novel insights on factors that effect siRNA potency

Motivation: Short interfering RNA (siRNA)-induced RNA interference is an endogenous pathway in sequence-specific gene silencing. The potency of different siRNAs to inhibit a common target varies greatly and features affecting inhibition are of high current interest. The limited success in predicting siRNA potency being reported so far could originate in the small number and the heterogeneity of available datasets in addition to the knowledge-driven, empirical basis on which features thought to be affecting siRNA potency are often chosen. We attempt to overcome these problems by first constructing a meta-dataset of 6483 publicly available siRNAs (targeting mammalian mRNA), the largest to date, and then applying a Bayesian analysis which accommodates feature set uncertainty. A stochastic logistic regression-based algorithm is designed to explore a vast model space of 497 compositional, structural and thermodynamic features, identifying associations with siRNA potency.

Results: Our algorithm reveals a number of features associated with siRNA potency that are, to the best of our knowledge, either under reported in literature, such as anti-sense 5′ −3′ motif ‘UCU’, or not reported at all, such as the anti-sense 5′ -3′ motif ‘ACGA’. These findings should aid in improving future siRNA potency predictions and might offer further insights into the working of the RNA-induced silencing complex (RISC).

Contact:cholmes@stats.ox.ac.uk

Supplementary information:Supplementary data are available at Bioinformatics online.

Chemical modification patterns compatible with high potency dicer-substrate small interfering RNAs

Dicer-substrate small interfering RNAs (DsiRNAs) are synthetic RNA duplexes that are processed by Dicer into 21-mer species and show improved potency as triggers of RNA interference, particularly when used at low dose. Chemical modification patterns that are compatible with high potency 21-mer small interfering RNAs have been reported by several groups. However, modification patterns have not been studied for Dicer-substrate duplexes. We therefore synthesized a series of chemically modified 27-mer DsiRNAs and correlated modification patterns with functional potency. Some modification patterns profoundly reduced function although other patterns maintained high potency. Effects of sequence context were observed, where the relative potency of modification patterns varied between sites. A modification pattern involving alternating 2'-O-methyl RNA bases was developed that generally retains high potency when tested in different sites in different genes, evades activation of the innate immune system, and improves stability in serum.

Modeling recursive RNA interference

An important application of the RNA interference (RNAi) pathway is its use as a small RNA-based regulatory system commonly exploited to suppress expression of target genes to test their function in vivo. In several published experiments, RNAi has been used to inactivate components of the RNAi pathway itself, a procedure termed recursive RNAi in this report. The theoretical basis of recursive RNAi is unclear since the procedure could potentially be self-defeating, and in practice the effectiveness of recursive RNAi in published experiments is highly variable. A mathematical model for recursive RNAi was developed and used to investigate the range of conditions under which the procedure should be effective. The model predicts that the effectiveness of recursive RNAi is strongly dependent on the efficacy of RNAi at knocking down target gene expression. This efficacy is known to vary highly between different cell types, and comparison of the model predictions to published experimental data suggests that variation in RNAi efficacy may be the main cause of discrepancies between published recursive RNAi experiments in different organisms. The model suggests potential ways to optimize the effectiveness of recursive RNAi both for screening of RNAi components as well as for improved temporal control of gene expression in switch off-switch on experiments.

siRNAs targeting mouse myostatin

Eight different mouse myostatin small interfering RNA (siRNAs) were synthesized and tested. Five siRNAs showed a pronounced biological effect reducing myostatin mRNA content. For two of them, the myostatin mRNA level was reduced 3- and 4-fold, respectively. The obtained siRNAs can be used for study of biological effects of myostatin, both in vitro and in vivo.

IDT SciTools: a suite for analysis and design of nucleic acid oligomers

DNA and RNA oligomers are used in a myriad of diverse biological and biochemical experiments. These oligonucleotides are designed to have unique biophysical, chemical and hybridization properties. We have created an integrated set of bioinformatics tools that predict the properties of native and chemically modified nucleic acids and assist in their design. Researchers can select PCR primers, probes and antisense oligonucleotides, find the most suitable sequences for RNA interference, calculate stable secondary structures, and evaluate the potential for two sequences to interact. The latest, most accurate thermodynamic algorithms and models are implemented. This free software is available at http://www.idtdna.com/SciTools/SciTools.aspx.