Published Paper
Download Model
GitHub Repository (legacy)
The links marked legacy point to the original published source code. For the model as this web server runs it, use the Download Model link above.
DeepPE is a web tool that designs prime editing (PE) guide RNAs (pegRNAs) and predicts their efficiencies and activities.
When a target sequence is entered, this tool provides the expected PE2 efficiencies for a total of 57 pegRNAs:
24 pegRNAs from DeepPE (deep-learning model)
DeepPE
23 pegRNAs from PE_type (random-forest model)
PE_type
10 pegRNAs from PE_position (random-forest model)
PE_position
Inputs that specify an edit. An unedited/edited sequence pair, ClinVar ID and COSMIC ID name the edit to install, and the pegRNA is designed to write it. No separate homology-directed repair (HDR) donor is offered on this workflow: in prime editing the pegRNA's own reverse-transcription template is the repair template, so a second single-stranded donor would be a different experiment rather than an extra column.
Note: DeepPrime and SynDesign are also available and may be used for a larger variety of prime editing guide RNA recommendations and efficiency predictions across diverse cell lines.