DeepPrime

Deep learning-based model that can predict the prime editing efficiency for all possible pegRNAs for a given target sequence

Published Paper Publication
DeepPrime

Webtool Usage

The webtool app can accommodate most applications using default parameters and using the most appropriate primed editing (PE) model for your experimental conditions. It can evaluate all possible prime editing guide RNAs (pegRNAs) for a given target according to the predicted the prime editing efficiency, DeepPrime score.

Analysis: Input Sequence

The webtool accepts 121 base pairs (bps) long wild-type and primed edited sequences where the right context sequence of the edit is adjusted corresponding to the size of the edit, as exemplified in the figure below. Example

Default parameters:
1) For PE system, we recommend using PE2max with optimized scaffold in HEK293T cells as the initial analysis. According to our findings, it provides the best general prediction performance across diverse target sites

2) Edit type and Edit length can be adjusted according to the experimental conditions.

3) Length ranges for the reverse transcription template (RTT) and primer binding site (PBS) can be set to max defaults for maximum number of pegRNA considerations. However, the ranges can be adjusted lower for faster processing times.




Analysis: ClinVar ID

The webtool can accept specific variant ID (or VCV ID) from the ClinVar database. Example



Analysis: Off-target
Currently, only the model trained on PE2 with conventional scaffold in HEK293T cells is currently available for running off-target prediction for specific pegRNAs. Even though this system can be limited in performance compared to the other optimized, fine-tuned models. The off-target analysis can provide insight into how prevalent off-target levels are.

Running off-target mode:
On the results page, use the checkbox indicating that you are currently running the off-target compatible analysis. Enter the pegRNA ID and off-target sequences (74bps) to evaluate.


Results:

After DeepPrime has processed the data, the Top Recommended pegRNA information is shown according to the oligonucleotide design parameters. General information for the pegRNA as well asach component for the oligo design of pegRNA has been streamlined as shown.
The rank plot shows the overall ranking of the top 5 pegRNA compared with the all DeepPrime scores from our model testing from each PE system. Download link at the end of each Results page can be used to obtain the full results anda all rank plots.