E-RNAi allows for the adjustment of several design and evaluation settings for long dsRNAs and siRNAs.
De novo design settings
Predicted specificity, efficiency and low complexity (long dsRNAs and siRNAs)
siRNA length for specificity prediction
This parameter defines the length of siRNAs used for off-target evaluation. Length between 16bp and 28bp can be selected. E-RNAi cuts the queried target sequence into all possible siRNAs of this length with an offset of 1, e.g. if the queried target sequence is 500bp long and a siRNA length of 19bp was selected the query gets cut into 482 siRNAs. These are then mapped (perfect matches only) to the transcriptome of the selected organism to assess their specificity using Bowtie.
Exclude regions of low sequence complexity (mDust)
The queried target sequence is filtered for regions of low complexity such as simple nucleotide repeats using the mDust filter. These regions are avoided for designs as they might lead to unwanted off-target effects.
Exclude >5x CA[ACGT] repeats
The queried target sequence is filtered for stretches containing more than 5 consecutive CA[ACGT] repeats (also called CAN repeats). Theses regions are avoided for designs as they were associated with unwanted off-target effects.
Prediction of siRNA efficiency
Set efficiency calculation method and efficiency cut-off ('Minimal siRNA efficiency score'). Available calculation methods are 'Rational' for calculations according to Reynolds et al. or 'Weighted' according to Shah et al. The efficiency cut-off defines the minimal required efficiency for an siRNA (contained in a long dsRNA) to be considered. A more detailed description of the efficiency prediction is available here.
Allow relaxation of parameters if required
In case the queried target sequence does not meet the selected quality cut-offs (specificity, low-complexity, CAN repeats, efficiency) these can be relaxed by E-RNAi during the design to allow the design of a reagent.
Mapping of designs to genome and homology evaluation (long dsRNAs and siRNAs)
Calculate homology of designs to transcripts using cut-off
The homology of the designed reagent to sequences of the transcriptome of the selected organism is calculated using BLAST. The cut-off value that can be defined is the BLAST E-value, which describes the likelihood that another sequence can occur in the transcriptome with a similar score just by chance. Evaluating the design for homology will show, whether the designed reagent shows homology to unintended target genes that might be targeted by imperfect matches.
Primer design options (long dsRNAs only)
These settings influence the length of the long dsRNA to be designed and the primer quality. primer3 is used for primer designs. Further documentation can be found there.
Optimal primer size (bp)
Optimum length (in bp) of a primer oligo. Primer3 will attempt to pick primers close to this length.
Minimal primer size (bp)
Minimum acceptable length (in bp) of a primer. Must be greater than 0 and less than or equal to selected 'Maximal primer size'.
Maximal primer size (bp)
Maximum acceptable length (in bp) of a primer.
Amplicon size range
These values specify the lengths of the dsRNA that the user wants the primers to create.
Maximal primer pair penalty
Maximum acceptable primer pair penalty. The lower it is the better the primers should work.
Number of primer designs per specific region
Number of primer pairs designed for a queried target sequence. Higher numbers slow down design process but increase probability to find an optimal sequence.
Addition of sequence tags to primers (long dsRNAs only)
Long dsRNAs are synthesized by in vitro transcription reactions. This option allows to add promoters required by T7 or SP6 RNA polymerases 5' to designed forward and reverse primers automatically. Instead of adding T7- or SP6- promoter sequences, an 'individual' sequence-tag can be defined.
Output options (long dsRNAs and siRNAs)
Number of designs reported per query sequence
This option defines the number of designs reported for each queried target sequence. If e.g. set to 5, the 5 best designs for the target sequence will be reported. These designs are not necessarily independent (the option for 'independent designs' described below can be used for that).
Independent designs (long dsRNAs and siRNAs)
This option allows to upload sequences (in FASTA format) that should be excluded (by homology) from the current design. E.g. in case a independent RNAi reagent (compared to a previously calculated one) should be designed the sequence of the first reagent can be uploaded here. E-RNAi then excludes this region from the current design.
Besides calculating independent reagents this option can also be used for other purposes e.g. for excluding qPCR amplicons from RNAi reagent sequences or for avoiding other unwanted sequences.
Further documentation about the required upload format (FASTA) can be found here.
All evaluation settings for the prediction of specificity, efficiency and low complexity as well as available output options are described above.
Reagent source type
For the evaluation of RNAi reagents the source that was used for the design of the reagent - either 'Genomic' (e.g. genomic DNA) or 'CDS' (e.g. cDNA) - can be selected. This influences the method E-RNAi uses to map the sequence of the reagent to the genome.
Analysis of siRNA pools
siRNAs are often used as pools. E-RNAi allows to evaluate siRNA pools and to generate a summarized report. This requires the upload of a tab-delimited file (with *.txt or *.tab extension) with two tab-separated columns containing the headers 'siRNAID' and 'POOLID'. 'siRNAID' is the sequence identifier of the siRNA (the FASTA header used to query the sequence) and 'POOLID' is the pool identifier, which must be the same for all sequence identifiers belonging to the same pool.
In the example below two pools were defined (M-003000-01 and M-003001-01). Each pool consists of 4 siRNAs (e.g. D-003000-05, D-003000-06, D-003000-07, D-003000-08 belong to the pool M-003000-01).