Note: SynDesign is also available with FASTA or gene input (1 bp substitution), enabling guide RNA design for saturation genome editing.
The Load Example button is able to replicate the example inputs shown in this help page for quick testing of the DeepPrime web server. Simply click the Load Example button then press Submit Job to obtain an example results page.
When this option is selected, this web server accepts wild-type (unedited) and prime edited sequences. The edit is read from the difference between the two, wherever it sits in them. 121 bp — 60 bp of flank on each side of the edit — offers the fullest set of candidate pegRNAs, because it is the shortest input in which every candidate's 74 nt scoring window fits; a shorter pair is designed from and offers fewer candidates.
Input Unedited Sequence(s) (121 bp recommended):
CTCACGTGAGCTCTTTGAGCTTGCCTGTCTCTGTGGGCTGAAGGCTGTTCCCTGTTTCCTTCAGCTCTACGTCTCCTCCGAGAGCCGCTTCAACACCCTGGCCGAGTTGGTTCATCATCAT
Input Edited Sequence(s) (121 bp recommended):
CTCACGTGAGCTCTTTGAGCTTGCCTGTCTCTGTGGGCTGAAGGCTGTTCCCTGTTTCCTACAGCTCTACGTCTCCTCCGAGAGCCGCTTCAACACCCTGGCCGAGTTGGTTCATCATCAT
CTCACGTGAGCTCTTTGAGCTTGCCTGTCTCTGTGGGCTGAAGGCTGTTCCCTGTTTCCTCTTTCAGCTCTACGTCTCCTCCGAGAGCCGCTTCAACACCCTGGCCGAGTTGGTTCATCAT
CTCACGTGAGCTCTTTGAGCTTGCCTGTCTCTGTGGGCTGAAGGCTGTTCCCTGTTTCCTAGCTCTACGTCTCCTCCGAGAGCCGCTTCAACACCCTGGCCGAGTTGGTTCATCATCATTC
When the above option is selected, the web server can accept a specific variant ID (or VCV ID) from the ClinVar database.
Four identifier forms reach the same variant: the numeric Variation ID (e.g., 9325), the matching VCV accession (e.g., VCV000009325.76), a ClinVar record (RCV) or submission (SCV) accession, and a dbSNP rs number (e.g., rs80357382). An RCV or SCV accession names an interpretation or a submission rather than the variant, and its digits are not the Variation ID, so it is looked up in the ClinVar release bundled with this server and translated to the Variation ID behind it; the job log states the translation used. An accession or rs number the bundled release does not carry is refused, with the reason.
9325
VCV000009325.76
rs80357382
ClinVar Analysis Mode determines whether the variant will be installed (Model) or corrected back to the reference (Correction) by the pegRNA design.
When this option is selected, the web server accepts a single mutation identifier from the Catalogue of Somatic Mutations in Cancer. Three forms reach the same mutation: a genomic mutation ID (COSV57468486), a legacy coding or non-coding ID (COSM476, COSN1234), and a bare number, which keeps its historical meaning of COSM. The prefix may be left off.
COSV57468486
COSM476
COSN1234
The identifier is resolved against the COSMIC release bundled with this server into the reference and alternate alleles at a genomic position, and the surrounding sequence is taken from the reference genome, so no sequence pair is asked for. One the bundled release does not carry is refused, with the reason.
ClinVar Analysis Mode applies here as well, and decides the same thing: Model designs pegRNAs that install the mutation, Correction designs pegRNAs that correct it back to the reference allele.
PE2max with optimized scaffold in HEK293T cells
Off-target prediction runs under the (Off-target compatible) PE2_Conv with conventional scaffold in HEK293T cells model alone. Even though this system can be limited in performance compared to the other optimized, fine-tuned models, the off-target analysis model can provide insights into how prevalent off-target levels are.
(Off-target compatible) PE2_Conv with conventional scaffold in HEK293T cells
This trained model is the whole of the off-target readout offered, here and on the other prime editors: there is no reference genome to choose, no genome-wide search, and no mismatch-tolerance score from a substitution table. Prime editors are far less sensitive to off-target editing than nucleases are — installing an edit needs the spacer to match, the nick to be made, the PBS to anneal and the RTT to be copied in, where a nuclease needs the spacer alone — so the sites a counted-mismatch genome search returns are not the sites a pegRNA edits. The model scores the sites named in the text area rather than finding them, and its per-group tables are their own download beside the off-target score column it adds to the results table.
Running off-target mode: With (Off-target compatible) PE2_Conv with conventional scaffold in HEK293T cells configured as the Prime Editor system, fill in the off-target sequences (74bp) in comma-separated format in the provided text area. You can run initially in on-target mode to find the top recommended pegRNA, and then use the 74bp wild-type (WT) sequence in the results page to derive the off-target sequences of interest.
The above example input is available after (1) pressing Load Example and then (2) configuring (Off-target compatible) PE2_Conv with conventional scaffold in HEK293T cells.
pegRNA Form decides the construct and is set to pegRNA (Unmodified 3' End) by default. Setting it to epegRNA appends a 3' structured motif to the extension and adds one orderable oligo pair per motif to the results table. The motif protects the pegRNA 3' end from exonucleolytic degradation. An epegRNA is a different construct to order from a pegRNA, which is why the motif is asked for rather than applied by default, and epegRNA Motif Panel — shown only while epegRNA is selected — chooses which motifs: the Recommended Panel (eHAV, tevo2.0, HAV), the Extended Panel (those three plus eSBRMV1-A, FMDVA-2 and tevopreQ1), Synthetic / LNP (eSBRMV1-A), the v5 Library Screen Panel (eHAV, tevo2.0, HAV, eSBRMV1-A, eFMDVA-2.2 and eFMDVO-4), or any single motif on its own. The panel is offered rather than ranked because no feature predicts which motif wins.
No linker is inserted between the primer binding site (PBS) and the motif, and that is deliberate rather than an omission. The v2 motifs offered here (tevo2.0, eHAV, HAV, eSBRMV1-A and the FMDV motifs) are used by Sakai et al. with the motif joined directly to the PBS 3' end; their assembled-construct tables carry no linker field. For tevopreQ1, Nelson et al. compared their pegLIT-designed 8-nucleotide linkers against no linker at all and reported no significant gain in editing efficiency (Supplementary Fig. 13). The one motif their data shows does need a linker is mpknot, which is not offered here.
If you are building an mpknot epegRNA elsewhere, design its linker with pegLIT (https://peglit.liugroup.us) rather than reusing these oligos.
Cloning Vector decides whether each candidate also carries an ordering-ready spacer oligo pair, and it is set to SpCas9 / lentiGuide-Puro (BbsI or BsmBI) by default. The pair arrives as Oligo (forward) and Oligo (reverse) in the results table and in the downloadable files alike: the two strands that anneal into the digested vector, CACC-presenting on the forward strand and AAAC-presenting on the reverse. Setting the vector to None (No Oligo Output) leaves both columns out rather than blank, and the four overhang fields override what the vector supplies.
Oligo (forward)
Oligo (reverse)
CACC
AAAC
Top Recommended pegRNA After DeepPrime has processed the data, the Top Recommended pegRNA information is shown according to the oligonucleotide design parameters. The table carries general information for the pegRNA alongside each component of its oligo design, so a candidate can be read in one row. All columns may be sorted and filtered in the table.
Alternatively, a slightly modified table is shown when the off-target model is run, which includes the DeepPrime score and off-target score.
The rank plot shows the overall ranking of the top pegRNA outputs compared with all DeepPrime scores from our model, testing from each PE system.
The results table shows the top pegRNAs and their major output columns.
Guide Sequence (Spacer): The 20-nucleotide spacer at the 5′ end of the pegRNA that directs the Cas9 nickase to its target. It is written as gN19 — a 5′ G in place of the protospacer's first base, which is the base the U6 promoter transcribes from, followed by protospacer positions 2 to 20.
PBS length (nt): The number of nucleotides in the PBS domain, which binds to the nicked DNA strand to initiate reverse transcription.
RT template length (nt): The length of the RTT domain, which contains the desired genetic edit to be integrated into the genome.
Edit type and length: The nature and size of the modification, read from the sequences themselves — sub1 a 1-bp substitution, ins3 a 3-bp insertion, del1 a 1-bp deletion.
sub1
ins3
del1
PBS-RTT (Reverse Complement): The combined nucleotide sequence of both the PBS and the RTT (due to matching the target sequence instead of the pegRNA, the actual oligonucleotide would be the reverse complement of this sequence).
DeepSpCas9 score (spacer activity): The nickase's predicted activity on this spacer, scored by DeepSpCas9 over its 30-nucleotide context. It measures how well the spacer engages, not how well the edit installs.
DeepPrime Score: The predicted prime editing efficiency (as a percentage) calculated by the DeepPrime deep-learning model for the specific pegRNA.
Z-Score: A normalized value used to standardize the predicted efficiency relative to the test data.
Score Percentile: The rank of the selected pegRNA's efficiency compared to all other possible pegRNA designs for the specific target edit.
Spacer / Extension cloning oligo (top and bottom strand): The DNA oligonucleotides to order for cloning. The spacer pair encodes the guide sequence and the extension pair the PBS and RTT domains. Cloning Vector decides whether the spacer pair is built at all — set it to None (No Oligo Output) and those columns are left out rather than blank — and the four overhang fields beside it (Forward 5′, Reverse 5′, Forward 3′, Reverse 3′) override the vector's own CACC/AAAC, as the four Extension fields do for the extension pair's GTGC/AAAA. Each field left blank takes the vector's own overhang.
GTGC
AAAA
Wild-type target sequence (74 nt): The unedited genomic sequence at the target site, in the 74-nucleotide window the model scores. An off-target sequence to compare against is derived from this one.
PBS-RTT with the edit marked (WT/edit): The PBS and RTT domains with the intended change written in place, each altered base shown as wild-type/edit. It is the pegRNA extension carrying the edit, not the edited genomic sequence.
The Download Results button at the end of each Results page can be used to obtain the full results and all rank plots.
The following example interactive outputs are available for reference:
If you have any questions, bug reports, or suggestions, please do not hesitate to contact us through the Support page.
© Laboratory of Genome Editing (Yonsei University)
