Prime edit mode fails with wrong strand

[slogemann1]

Hello,
first of all, thank you very much for your work on CRISPResso. Recently I used CRISPResso (v2.3.3) to evaluate prime editing data from paired-end reads from 156bp MiSeq sequencing data. I used prime editing mode and provided all parameters for the pegRNA including spacer sequence, scaffold, and extension. Based on my primers for creating the library the sequenced amplicon after merging is the reverse compliment of the expected input. CRISPResso failed even when providing the option --prime_editing_override_sequence_checks

Expected behavior
I would expect either a more informative error message or information in the documentation since I only discovered that there is no attempt to check against the reverse compliment of the amplicon by looking through the source code.

To reproduce
Take prime editor amplicon sequencing data with paired-end reads that has the reverse compliment of the expected sequence. You should be able to generate such a dataset by running seqtk seq -r on both files of correct orientation data, since that was ultimately my solution to fix this error.
Run CRISPResso as follows (example amplicon left for clarity):

CRISPResso -r1 r1.fq.gz -r2 r2.fq.gz --amplicon_seq "GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG" -p 10 --prime_editing_pegRNA_spacer_seq "GCGGGGGTACATCTCCATTC" --prime_editing_pegRNA_extension_seq "TCGTGTATCACAGGAATGGAGATGTA" --prime_editing_pegRNA_scaffold_seq "GTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGC" --prime_editing_override_prime_edited_ref_seq "GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACAGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG"

Debug output

INFO  @ Thu, 08 Jan 2026 16:20:37 (0.0% done):

                                                   ~~~CRISPResso 2~~~
                            -Analysis of genome editing outcomes from deep sequencing data-

                                                            _
                                                           '  )
                                                           .-'
                                                          (____
                                                       C)|     \
                                                         \     /
                                                          \___/

                                               [CRISPResso version 2.3.3]
[Note that as of version 2.3.0 FLASh and Trimmomatic have been replaced by fastp for read merging and trimming. Accordingly, the --flash_command and --trimmomatic_command parameters have been replaced with --fastp_command. Also, --trimmomatic_options_string has been replaced with --fastp_options_string.

Also in version 2.3.2, when running CRISPRessoPooled in mixed-mode (amplicon file and genome are provided) the default behavior will be as if the --demultiplex_only_at_amplicons parameter is provided. This change means that reads and amplicons do not need to align to the exact locations.]
                           [For support contact k.clement@utah.edu or support@edilytics.com]


WARNING @ Thu, 08 Jan 2026 16:20:37 (0.0% done):
         Folder CRISPResso_on_r1_r2 already exists.

DEBUG @ Thu, 08 Jan 2026 16:20:37 (0.5% done):
         CRISPRessoPro not installed


INFO  @ Thu, 08 Jan 2026 16:20:37 (0.5% done):
         pegRNA spacer alignment:
Forward (correct orientation):
-----------------------------------------------------------------------------GCGGG----GGTACATCTCCATTC-----------------------------------------
GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG
Score: 9.155
Reverse (incorrect orientation):
---------------------------------------------------------------GCGGGGGTACATCTCCATTC-----------------------------------------------------------
CTCATCCTCCTCGGCCAGCCTGCCTTCCTCCGGCAGGAGCAGCCTGGGCAGTCACCAGCTCCCGCGGGGGTACATCTCCATTCCGGTGATACACGAGCAGAACGTTACCCGGCCAGCAGCCCAGCCCTCCTTCCACCAAGCC
Score: 14.085

Traceback (most recent call last):
  File "/mnt/c/Users/user/Desktop/wsl/Crispresso2/python-venv/lib/python3.12/site-packages/CRISPResso2/CRISPRessoCORE.py", line 2855, in main
    raise CRISPRessoShared.BadParameterException(error_msg)
CRISPResso2.CRISPRessoShared.BadParameterException: The prime editing pegRNA spacer sequence appears to be given in the 3'->5' order. The prime editing pegRNA spacer sequence (--prime_editing_pegRNA_spacer_seq) must be given in the RNA 5'->3' order.
CRITICAL @ Thu, 08 Jan 2026 16:20:37 (0.5% done):
         Traceback (most recent call last):
  File "/mnt/c/Users/user/Desktop/wsl/Crispresso2/python-venv/lib/python3.12/site-packages/CRISPResso2/CRISPRessoCORE.py", line 2855, in main
    raise CRISPRessoShared.BadParameterException(error_msg)
CRISPResso2.CRISPRessoShared.BadParameterException: The prime editing pegRNA spacer sequence appears to be given in the 3'->5' order. The prime editing pegRNA spacer sequence (--prime_editing_pegRNA_spacer_seq) must be given in the RNA 5'->3' order.


CRITICAL @ Thu, 08 Jan 2026 16:20:37 (0.5% done):
         Parameter error, please check your input.

ERROR: The prime editing pegRNA spacer sequence appears to be given in the 3'->5' order. The prime editing pegRNA spacer sequence (--prime_editing_pegRNA_spacer_seq) must be given in the RNA 5'->3' order.

I was unable to find any option in the CRISPRESSO documentation that would let me get around this error, since the --prime_editing_override_sequence_checks option also fails when looking at the extension. I would suggest as a simplest solution to clarify what exactly the prime editor mode does in the documentation and that it expects a certain orientation and perhaps link to a solution with seqtk processing. Alternatively I would suggest an additional flag to take the reverse compliment of the merged sequences before processing or automatically checking and accounting for this without additional user input but writing it to the console as a warning. In case I have time I could try working on it myself if you see any of these as viable options.

Hopefully I did not oversee any solutions or make any other mistakes, otherwise please let me know. Thank you very much for your work!

[kclem]

Hi @slogemann1,

Thanks for your thorough comment. In regular CRISPResso mode i.e. for Cas9 analysis, we allow users to provide the amplicon sequence in the forward or reverse direction. However, because the editing outcomes and experimental designs for prime editing are much more complex than Cas9, we require users to provide the reference and prime editing input sequences in a standardized orientation.

The error message "ERROR: The prime editing pegRNA spacer sequence appears to be given in the 3'->5' order. The prime editing pegRNA spacer sequence (--prime_editing_pegRNA_spacer_seq) must be given in the RNA 5'->3' order." indicates that your sequence is provided in the incorrect orientation. Unfortunately, without knowing more details about your experimental design I'm not sure what the correct sequences for your analysis are.

Check out this example with a schematic and the correct orientations of prime editing parameters: #197

Try running CRISPResso with the parameters:
-a the WT amplicon sequence
--prime_editing_pegRNA_spacer_seq in RNA 5'->3' order, so for Cas9, the PAM is on the right side
--prime_editing_pegRNA_extension_seq RNA 5'->3' order, so in the order of RT>edit>PBS
--prime_editing_pegRNA_scaffold_seq in RNA 5'->3' order.
--prime_editing_nicking_guide_seq in RNA 5'-3' order, so for Cas9 the PAM is on the right side.

In early versions of CRISPResso, we attempted to infer the correct orientation of the amplicon and pegRNA components automatically. However, this became unreliable, especially for prime editing experiments involving large insertions or deletions, so we now require users to specify all sequences explicitly in the orientations described above. These conventions should also match the orientation used when ordering reagents. At this time, we don’t plan to revert to automatic orientation inference.

I hope this helps, and let me know if you still run into errors when providing the parameters as described above.

[slogemann1]

Hello there!
Thank you very much for your response. For context I will add my full epegRNA sequence taken from the plasmid I use.

5' CACC             GCGGGGGTACATCTCCATTC GTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGC TCGTGTATCACAGGAA TGGAGATGTA CGCGGTTCTATCTAGTTACGCGTTAAACCAACTAGAA TTTTTTT    3'
5' ... U6 promoter  protospacer          gRNA scaffold                                                                RTT              PBS        extension element                     terminator 3'

Here is also the alignment with the reverse complimented amplicon in 5' to 3' direction (using the reverse compliment of the protospacer)

                                                                  GCGGGGGTACATCTCCA TTCPAM
5' CTCATCCTCCTCGGCCAGCCTGCCTTCCTCCGGCAGGAGCAGCCTGGGCAGTCACCAGCTCCCGCGGGGGTACATCTCCA TTCCGGTGATACACGAGCAGAACGTTACCCGGCCAGCAGCCCAGCCCTCCTTCCACCAAGCC 3'

Here is the amplicon from my unchanged sequencing data

GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG

Based on this my sequences should be in the right direction, shouldn't they?

Just for certainty, I also reran the sequences with the pegRNA components reversed since this should correct it if they were really given in 3' to 5' direction.
Here is the command:

/usr/local/bin/CRISPResso -r1 r1.fq.gz -r2 r2.fq.gz --amplicon_seq GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG -p 10 --prime_editing_pegRNA_spacer_seq CTTACCTCTACATGGGGGCG --prime_editing_pegRNA_extension_seq ATGTAGAGGTAAGGACACTATGTGCT --prime_editing_pegRNA_scaffold_seq CGTGGCTGAGCCACGGTGAAAAAGTTCAACTATTGCCTGATCGGAATAAAATTGAACGATAAAGATCGAGATTTTG --prime_editing_override_prime_edited_ref_seq GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACAGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG --debug

And here is the output:

INFO  @ Fri, 09 Jan 2026 13:27:48 (0.0% done):

                                                   ~~~CRISPResso 2~~~
                            -Analysis of genome editing outcomes from deep sequencing data-

                                                            _
                                                           '  )
                                                           .-'
                                                          (____
                                                       C)|     \
                                                         \     /
                                                          \___/

                                               [CRISPResso version 2.3.3]
[Note that as of version 2.3.0 FLASh and Trimmomatic have been replaced by fastp for read merging and trimming. Accordingly, the --flash_command and --trimmomatic_command parameters have been replaced with --fastp_command. Also, --trimmomatic_options_string has been replaced with --fastp_options_string.

Also in version 2.3.2, when running CRISPRessoPooled in mixed-mode (amplicon file and genome are provided) the default behavior will be as if the --demultiplex_only_at_amplicons parameter is provided. This change means that reads and amplicons do not need to align to the exact locations.]
                           [For support contact k.clement@utah.edu or support@edilytics.com]


WARNING @ Fri, 09 Jan 2026 13:27:48 (0.0% done):
         Folder CRISPResso_on_r_1_r_2 already exists.

DEBUG @ Fri, 09 Jan 2026 13:27:48 (0.5% done):
         CRISPRessoPro not installed


INFO  @ Fri, 09 Jan 2026 13:27:48 (0.5% done):
         Alignment between extension sequence and reference sequence:
--------------------------------------------------------------------------------------------------------------------------------------------AGCACATAGTGTCCTTACCTCTACAT
GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG------------------------

INFO  @ Fri, 09 Jan 2026 13:27:48 (0.5% done):
         Using provided prime_editing_override_prime_edited_ref_seq
Detected: GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAGCACATAGTGTCCTTACCTCTACAT------------------------
Provided: GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACAGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG

Traceback (most recent call last):
  File "/mnt/c/Users/user/Desktop/wsl/Crispresso2/python-venv/lib/python3.12/site-packages/CRISPResso2/CRISPRessoCORE.py", line 2882, in main
    raise CRISPRessoShared.BadParameterException('The provided extension sequence is not in the prime edited override reference sequence!')
CRISPResso2.CRISPRessoShared.BadParameterException: The provided extension sequence is not in the prime edited override reference sequence!
CRITICAL @ Fri, 09 Jan 2026 13:27:48 (0.5% done):
         Traceback (most recent call last):
  File "/mnt/c/Users/user/Desktop/wsl/Crispresso2/python-venv/lib/python3.12/site-packages/CRISPResso2/CRISPRessoCORE.py", line 2882, in main
    raise CRISPRessoShared.BadParameterException('The provided extension sequence is not in the prime edited override reference sequence!')
CRISPResso2.CRISPRessoShared.BadParameterException: The provided extension sequence is not in the prime edited override reference sequence!


CRITICAL @ Fri, 09 Jan 2026 13:27:48 (0.5% done):
         Parameter error, please check your input.

ERROR: The provided extension sequence is not in the prime edited override reference sequence!

This also does not work, yet taking the reverse compliment of all the sequences and then using the reverse compliment as a reference sequence works perfectly. All the sequences are then also properly aligned in the output document. For this reason I think it is not a problem of the pegRNA parameters but rather the orientation of the sequencing reads.

Assuming this is correct then so far I don't see any indication that this is the problem I should have been looking for in the error messages or in the documentation on the official website.

Thank you very much for your help. Let me know if this makes sense!

[kclem]

I still can't see where your extension sequence aligns to your reference sequence.

I can see that your PBS aligns to your reference, but normally the reverse transcription template aligns with your reference after the programmed edit - I don't see this in yours. Could you help me understand how you expect this to look in your experiment?

PBS + extension                                                        TGGAGATGTACGCGGTTCTATCTAGTTACGCGTTAAACCAACTAGAA
                                                                       |||||||||||
amp_ref  GGCTTGGTGGAAGGAGGGCTGGGCTGCTGGCCGGGTAACGTTCTGCTCGTGTATCACCGGAATGGAGATGTACCCCCGCGGGAGCTGGTGACTGCCCAGGCTGCTCCTGCCGGAGGAAGGCAGGCTGGCCGAGGAGGATGAG
                                                                    ||||||||||||||||||||
gRNA rc                                                             GAATGGAGATGTACCCCCGC