• pssRNAMiner
• psRNATarget
• pssRNAit

pssRNAit: Designing Effective and Specific Plant RNAi siRNAs with Genome-wide Off-target Gene Assessment    Link

Limitations of past methods: Gene silencing through RNA interference (RNAi) is a widely used molecular tool in plants and animals for functional genomics. The biogenesis of siRNA and its binding to the target for gene silencing is multi-step process of RNA interference (RNAi) pathways. Although a number of siRNA design tools have been developed, however, it is still challenging to design effective, specific and non-toxicity siRNAs against a target gene particularly for plants. The induction of RNAi in plant cell is mainly accomplished by expressing 300-1,200 bp Gene-Specific Sequence Tag (GST) that makes long dsRNA which process through Dicer-like (DCL) enzyme to generate small interfering RNAs (siRNAs) (1). However, siRNAs generating from GST can be a mixture of effective, non-effective, toxic, non-toxic, specific, and non-specific small RNAs, which poses a great challenge during gene silencing experiments in term specificity and effectiveness, and this still was not addressed.

Novelty: We present pssRNAit, a web server tool to design effective, specific and non-toxic siRNAs for plant RNAi. This tool implemented several innovative approaches based upon recent understanding of biological mechanism of RNAi pathways gain through our findings and literatures. For this, we have developed reliable computational models specific to each step of RNAi pathways and integrated these models which works like pathways to design siRNAs.

pssRNAit integrated several models and cDNA transcripts library.

1. A new SVM model is use to design highly effective siRNA.
2. Remove siRNA which contains toxic and non-specific sequence motifs (2).
3. Our RISCbinder model is use to select those effective siRNA (antisense:sense) whose antisense strand have binding affinity with RISC machinery to execute gene silencing (3).
4. Our psRNATarget tool is used to predict off-target genes of design siRNAs and to select more specific siRNAs (4).
5. Our recent finding using high-throughput data of miRNAs showed several isomiRs are also generating along with canonical miRNA in order to increase the target specific gene silencing (5). Therefore, we are intelligently select pool of siRNAs to further increase the specificity of gene silencing. The basic principle is to select bunch of siRNAs which have common target gene but different off-targets.

Input/Output: The server front-end integrates simplified user-friendly interfaces to accept mRNA/cDNA sequence in FASTA format as input. The species name should be chosen to automatically load the cDNA/transcript libraries of this species for genome wide off-target assessment. The input interface also has options to remove siRNA containing toxic and non-specific sequence motifs. Upon submission, its backend pipeline that runs on linux cluster designs best siRNAs and gives the output result as table contains antisense and sense sequence of siRNA, alignment of siRNA binding with users’ sequence, silencing efficiency, number of off-targets and details through a link.

Link