CRISP-ID: Detecting CRISPR mediated indels by Sanger sequencing.

CRISP-ID is a web application that allows the detection of the exact indel size and location of a CRISPR-Cas9 targeted region, based on direct Sanger sequencing.

Watch the introduction video to get started:


Please load a Trace file

Reference sequence

Loading...

Enable colors:

Citation


If you use CRISP-ID, please cite:

Dehairs, J. et al. CRISP-ID: decoding CRISPR mediated indels by Sanger sequencing Sci. Rep. 6, 28973; doi: 10.1038/srep28973 (2016).

Link to original article: http://www.nature.com/articles/srep28973

Example data


An example dataset can be downloaded here.

FAQ


I'm getting an error message saying that my trace file is not valid.

Users have reported incompatibility with ab1 files generated by the latest generation of Sanger sequencers. We are working on support for these kinds of ab1 files. Meanwhile if you experience problems with your files, we recommend using the SCF format instead. The freely available Snapgene Viewer can be used to convert ab1 files to SCF.


I'm getting the message "Warning! Couldn't find alignment with the reference sequence."

Please ensure that you have entered a valid reference sequence. It is best practice that the reference sequence approximately matches the sequenced region in length. Providing a very short reference sequence (<100 bp) or an unnecessary long one (>1000 bp) may cause problems. If you have provided an appropriate reference sequence, this error usually occurs when the quality of the spectrum is poor. You might have to adjust the background cutoff to improve the accuracy of the base calling, or remove problematic regions by adjusting the start and end positions. Also note that if you are using a reverse sequencing primer, you have to supply the reverse complement of the reference sequence.


Can CRISP-ID analyze files with 4 or more overlapping traces?

CRISP-ID does not support mixed sequences with more then 3 traces. While it is theoretically possible to handle more then 3 alleles, the more traces that are present, the longer the sequence run will have to be for CRISP-ID to find the correct frame for all traces (because homozygous base calls become very unlikely). Combined with the technical reality that the higher the trace number, the more the quality of the spectrum degrades (height of the peaks decreases, peaks no longer align,...) these problems make it unfeasible to resolve more than 3 sequences.


Can CRISP-ID be used to find indels other than those induced by CRISPR/Cas9?

CRISP-ID can be used to look for any kind of indel on up to 3 different alleles. It cannot however handle more then one indel per allele (e.g. multiple indels spaced apart).


How does CRISPR-ID deal with multiple random insertions of the same length at the same locus?

In case of this rare event it is not possible to determine the exact sequences at this location. The size and location of the insertion however will be reported correctly, which should be sufficient information if achieving a knock out is the goal.


Can CRISPR-ID be used if I'm using a repair template to introduce a specific modification?

When a repair template is provided to the cells and in the event that this results in a mixed spectrum, 2 different situations can be considered:
1. The CRISPR/Cas9 has induced a double strand break, but the homology-directed repair mechanism failed, resulting in non homologous end joining. This is a standard case for CRISP-ID.
2. The repair template has integrated on 1 allele but the other allele is wild type or has an indel. For CRISP-ID, this situation is really no different from the first and can be resolved. Be aware though, that in rare cases, if the repair template introduces an insertion on one allele, and the other allele has a NHEJ mediated random insertion in the same location, the bases in that location might not be assigned to the correct allele (cf previous question). This problem can be solved by providing a reference sequence that contains the template mediated insertion.


Do my sequences really have all these random mutations?

CRISP-ID was designed to determine the size and locus of indels and does this by trying to determine the exact sequences as accurately as possible. If the spectrum is of poor quality, or one of the signals is significantly weaker then the other, the accuracy of the base calling can be compromised. It is recommended to always take a moment to review the quality of the spectrum and the base calls. Poor performance may be remedied by trimming the sequence or adjusting the % background cutoff value, as illustrated in the example below:
background_cutoff

Changelog


CRISP-ID V1.1a (released 30-03-2017)
This is a minor update. The following minor new features were added:


CRISP-ID V1.1 (released 17-07-2016)
This is a minor update. The following minor new features were added:

Contact


If you have any questions or problems related to CRISP-ID, please let us know: