Go to this page to design your custom CRISPR guide RNAs.
Design of CRISPR constructs
Design Evaluation MultiCRISP CLD GenomeCRISPR Help Links

Check out our new CRISPR Library Designer (CLD): batch design of sgRNA libraries
Download the dockerized version now at CLD on Github


Further information about our Data Protection Policy can be found here: Privacy Policy


Florian Heigwer, Marco Breinig, Tianzuo Zhan, Michael Boutros
E-mail: crispr@dkfz.de

Former Contributors

Programming: Grainne Kerr, Oliver Dreier, Johanna Kratzer, Pauline Burkhardt, Alexander Mattausch, Pelz Oliver
Ideas: Marco Breinig, Tianzuo Zhan, Jan Winter, Dirk Brüggemann

How to cite

Heigwer, F. , Kerr, G. & Boutros, M. E-CRISP: fast CRISPR target site identification. Nat. Methods 11, 122-123 (2014).


E-CRISP is an online tool to design and evaluate Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)

E-CRISP has been optimized using fast and accurate algorithms to design CRISPR gRNA sequences to target any nucleotide sequence ranging from single exons to entire genomes. Special emphasis in the design process has been given to usability in experimental applications. E-CRISP not only checks for target specificity of the putative designs but also assesses their genomic context (e.g. exons, transcripts, CpG islands).


The input options are divided into 6 sections, each dealing with a different aspect of the design process
Section Aspect
Organism The organism the designs should be created for. The databases are pre-built for each organism. The genome release is indicated in the dropdown menu.
Target Sequence The sequence the CRISPR should be designed to target. Either enter an Ensembl ID, a gene symbol or a sequence in fasta format. If a fasta sequence is given the locus can be stated in the header in the form of "chrom:X:1..1000 ", if your sequence originates from the first 1000 bases of the X chromosome. If this location is not be stated in the header (the text after ">") the program does not check the genomic context.
Design Purpose In this section, the user can specify the experimental purpose of the CRISPR. Depending on the purpose, different regions of the input sequence will be targeted. Purposes included, knock-out experiments, N-Terminal tagging, C-Terminal tagging, CRISPRi and CRISPRa.
Further many more specific parameters can be tuned there. Such as favourable GC-content or PAMs.

Purpose Rules
  • target up to the first x defined coding exons
  • target only coding sequences of the gene of interest
  • take care of designs beeing downstream the start codon
N-terminal tagging
  • target needs to overlap with a certain windows around the start codon
    can be viewed by selecting to show start codons
C-terminal tagging
  • target needs to overlap with a certain windows around the stop codon
    can be viewed by selecting to show stop codons
  • target needs to overlap with a certain defined window around the TSS sites of that gene
    can be viewed by selecting to show TSS
  • target needs to overlap with a certain defined window around the TSS sites of that gene
    can be viewed by selecting to show TSS
Gene annotation filtering In this section, the user can filter the output results, based on gene annotation information. For example, all results which do not target an exon can be excluded from the output, or the user can specify which exon to target.
Off-target Analysis In this section, the user can specify parameters to search for off-target effects (regions where the design targets outside of input query sequence)
Output In this section, the user can specify what output files are produced. If the user expects a lot of CRISPR designs to be return (e.g. inputting a large number of sequences at once), the user can switch off producing an image and an html output table.


A summary of the design process.
  • Total number of possible designs: The total number of designs found for the input sequence
  • Number of successful designs: The number of designs returned as results
  • Numbers of CRISPR designs excluded and not returned:
    because they were not directly behind the start codon
    because they were to unspecific
    because they were not amaneable for CRISPRa/i
    because they were located in an CpG island
    because they did not hit any exon
    because their nucleotide composition was too invariable
A html table is returned, where each row indicates a CRISPR alignment.
Column Meaning
Name The ID of the CRISPR. This is of the form: ID of the input sequence_randomNumber_random_Numer.
Nucleotide Sequence The sgRNA target sequence
SAE Score S: Specificity score A: Annotation score E: Efficiency score. See the table below for more information.
Target The gene that is targeted by this gRNA. If a fasta sequence is given as input with no chromosome location information, E-CRISP cannot search the annotation databases, and no target gene will be returned.
Match String A coloured match screen, which indicates at a glance how good the alignment is: A green "M" for a match. A "X" for a mismatch, an "I" for an insertion in the gRNA.
Number of Hits The number of locations this CRISPR design targets, or, the number of times this CRISPR appears in the output table (one for each target).

A genome browser image of the CRISPR designs in their genomic context. This allows the user to visually inspect where the CRISPR in the input sequence. Off-targets (where the CRISPR targets outside of the input sequence) are not shown in this image.

There is also an option to output a gff file (http://www.ensembl.org/info/website/upload/gff.html)

The output table is available as tab-delimited file *.tab or Excel compatible file *.xls. And contains the following columns:
Column Meaning
Nametarget site ID
Lengthtarget length
Starttarget start with respect to the input sequence (gene -500 if a gene name was the input)
Endtarget start with respect to the input sequence (gene -500 if a gene name was the input)
Strandstrand it will target
Nucleotide sequencetarget site nucleotide composition of the form target_PAM
Gene NameID::GENE
TranscriptsENSEMBL transcript Ids overlapping with the target site
Transcript:: ExonENSEMBL transcript::exon Ids overlapping with the target site
Number of Cpg Islands hitNumber of Cpg Islands overlapping with the target site
Sequence around the cutsideif it is chosen to save a recombination matrix ist sequence is here
%A %C %T %Gnucleotide compositions in per cent
S-Scorespecificity score
A-Scoreannotation score
E-Scoreefficacy score
percent of total transcripts hitper cent of transcripts of the targeted gene being hit by that putative sgRNA
Targettarget genes by remapping the target site
Match-startalignment start with respect to the estimate target gene
Match-endalignment end with respect to the estimate target gene
Matchstringalignment representation "M" for match "X" for mismatch "I" for insertion "D" for deletion
Editdistanceestimated edit distance of the alignment (X+I+D)
Number of Hitsestimate number of target sites in the respective genome with the off-target parameters specified
Directionstrandedness of the target alignment
CDS_scorestart with 0. for every CDS of every transcript the target site ovelaps with add 5/CDS_number
Exon_Scorestart with 0. for every exon of every transcript the target site ovelaps with add 5/exon_number
seed_GCGC content in per cent of the 8 PAM proximal basepairs
Doench_ScoreEfficacy score as introduced by Doench et al. 2014 Nat. Biotech.
Xu_scoreEfficacy score as introduced by Xu et al. 2015 Gen.Res.
Chromosomechromosome the sgRNA targets
Genomic startgenomic location on that chromosome the sgRNA is targeting
Genomic Endgenomic location on that chromosome the sgRNA is targeting

Scoring: All scores given are normalized to 100 % reachable score

In addition the scores of Doench et al. and Xu et al. are given in the output table.
Specificity Score (S-score) Annotation Score (A-score) Efficacy Score (E-score)
Start with 100
for every off-target substract (20-mismatches)/iteration
Start with zero
For every hit exon add 5/exon count
For every hit CpG Island subtract 1
For every start codon hit add 1
For every stop codon hit add 1
For every CDS hit add 5/CDS count
For every gene hit add 1
Add 1 if the the last 6 bp have a CG content higher then 70 %
Subtract 1 if the entire sequence has GC content > 80 %
Add 1 if sequence is preceded by a G
Add 1 if there are GG in front of the target sequence (opposite the PAM)
Add micro -homology score (is higher when sequence tends to give out of frame deletions)

Frequently asked questions:

Why do I have to choose the organism for my design?

E-CRISP not only identifies if your input sequence has a CRISPR target site, it also annotates this site with genomic annotation information, such as which gene, transcript, exon are that the targeted site, if any. It also, check for off targets in the rest of the mRNA, transcripts or chromosomal DNA of that organism. In order to do this, you must select an organism, so that the correct genomic annotation databases and off-target databases can be searched.

How does E-CRISP annotate my sequence?

Every putative binding site found is annotated with its genomic context, i.e. whether it is contained within an exon, coding sequence, transcript, gene, CpG island etc. Annotation of many putative bindings sites requires an efficient search of genome annotation databases. To maximize efficiency and shorten runtime, E-CRISP uses a binary interval which stores all genome annotations for the respective organism.

Binary Tree

How does E-CRISP identify off-targets?

Why do I have a choice against which sequence to check for off-targets?

How do I identify if my CRISPR design has any off-targets from the output table?

What is a secondary off-target?

Check if the CRISPR targets any foreign, exogenously introduced sequences. You can select from a list of commonly introduced sequences in lab, or paste in the sequence in the text area.

I only get a chromosome name in the target column of the output table - why?

E-CRISP can only return target information, if location information is given in the input. If a gene name/symbol is given, this location information can be retrieved from the pre-built databases. If a fasta sequence is given, the location must be given in the fasta header, in order to check for genomic context.

If it is an off-target match, this match may lie outside an annotated region, in which case only the chromosome is returned.

Why is the minimum/maximum guide RNA length after PAM 20?

Recent publication have shown that the guide RNA target sequence might be as well shorter or longer than 20 bp. This can have influence on the binding affinity and thus the efficiency of the CRISPR construct.

1. Gasiunas, G. & Siksnys, V. RNA-dependent DNA endonuclease Cas9 of the CRISPR system: Holy Grail of genome editing? Trends Microbiol. (2013). doi:10.1016/j.tim.2013.19.001

How does E-CRISP work?

Where is the information about the organisms taken from?

Change Log:

2 Jul 2019, version 5.4 1. An option to exclude homopolymer stretches of nucleotides has been added.

2. A bug, whereby homopolymers were only searched in forward direction has been removed.

3. Stringency levels were adapted such that strict parameters are less strict and all respect the homopolymer exclusion

4. Homo sapiens became the default organism.

12 Jul 2016, version 5.1 1. Again new organisms including Lifestock, crops, funghi and bacteria have been added to E-CRISP

2. All other databses were updated to the newest version available in ensembl.

3. TSS, start codons and stop codons can be vizualized in the results

4. Scoring schemes of Doench et al. and Xu et al. added to the analysis.

5. Genomic location added to the output

6. Excel support added to the output

7. Output image beatified

8. fixed some minor bugs regarding the strandedness of genes and thus tagging results

9. implmented,tested and fixed rulesets for CRISPRa and CRISPRi

10. fixed flexible PAM input
22 Jan 2015, version 4.2 1. Many new organisms including Lifestock, crops, funghi and bacteria have been added to E-CRISP

2. All other databses were updated to the newest version available in ensembl.

3. Annotations have to overlap with the pam and not only some part of the sgRNA target

4. Targets can be identified in sequences which do not originate from the organism selected.

5. CpG islands are again properly shown in the result image
01 August 2014, version 4.0 E-CRISP has been reworked to inlcude the latest scientific results of the last months:

1. The following organisms hav been added:
Toxoplasma gondii GT1 (ToxoDB-10.0)
Gasterosteus aculeatus (Three-spined stickleback, BROADS1.75)
Populus trichocarpa (Black cottonwood, JGI2.0.21)
Sus scrofa (Pig, Sscrofa10.2.75)

2. A new more intuitive scoring system, devided into Specificity, Annotation and Efficiency score has been implemented.
Design results are sorted by Specificity, then Annotation and then efficiency

3. Three new default options have been added guiding you fastly to the most wanted results.
For further details visit the help pages and scroll down to the schoring scheme.

4. Off-target checks are now much more precise, because the PAM region (NAG or NGG) now is truely ambigous.
An off-target is searched without the PAM but only considered valid if any PAM is present.
26 May 2014, version 3.1 We are happy to announce a further major update to our E-CRISP web service.
Many new organisms have been added together with big changes in the web front end.
Hence you will find the new forum and many other new things here in the new BETA version 3.1.
14 April 2014, version 3.0.2 In this minor update different default values for de-novo sgRNA design have been implemented, allowing for more designs to be found.
01 April 2014, version 3.0.1 The following organisms have been added to E-CRISP:
Zea mays
Ustilago hordei
Toxoplasma gondii ME49
Gasterosteus aculeatus
Populus trichocarpa
20 March 2014, version 3 A new version of E-CRISP has been released (version 3.0). It includes more off-target search options and we implemented speed improvements to enable the design of sgRNAs against up to 200 genes in parallel.


1. Wang, T., Wei, J. J., Sabatini, D. M. & Lander, E. S. Genetic screens in human cells using the CRISPR-Cas9 system. Science 343, 80–4 (2014).
2. Fu, Y. et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat. Bio31, 1–6 (2013).
3. Malina, a. et al. Repurposing CRISPR/Cas9 for in situ functional assays. Genes Dev. 27, 2602–2614 (2013).
4. Ma, M., Ye, A. Y., Zheng, W. & Kong, L. A guide RNA sequence design platform for the CRISPR/Cas9 system for model organism genomes. Biomed Res. Int. 2013, 270805 (2013).
5. Smith, C. et al. Whole-Genome Sequencing Analysis Reveals High Specificity of CRISPR/Cas9 and TALEN-Based Genome Editing in Human iPSCs. Cell Stem Cell 15, 12–13 (2014).
6. Yu, Z. et al. Highly efficient genome modifications mediated by CRISPR/Cas9 in Drosophila. Genetics 195, 289–91 (2013).
7. Cho, S. W. et al. Analysis of off-target effects of CRISPR/Cas-derived RNA-guided endonucleases and nickases. Genome Res. (2013).
8. Shalem, O. et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343, 84–7 (2014).
9. Rna, C. S. pyogenes. 12, 1–7 (2013).
10. Mali, P. et al. CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering. Nat. Biotechnol. 31, 833–8 (2013).
11. Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–6 (2013).
12. Baena-Lopez, L. A., Alexandre, C., Mitchell, A., Pasakarnis, L. & Vincent, J.-P. Accelerated homologous recombination and subsequent genome modification in Drosophila. Development (2013).
13. Hou, Z. et al. Efficient genome engineering in human pluripotent stem cells using Cas9 from Neisseria meningitidis. Proc. Natl. Acad. Sci. 1313587110– (2013).
14. Hsu, P. D. et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nat. Biotechnol. 31, 827–32 (2013).
15. Jiang, W. et al. Demonstration of CRISPR/Cas9/sgRNA-mediated targeted gene modification in Arabidopsis, tobacco, sorghum and rice. Nucleic Acids Res. 1–12 (2013).
16. Ran, F. A. et al. Double Nicking by RNA-Guided CRISPR Cas9 for Enhanced Genome Editing Specificity. Cell 154, 1380–9 (2013).
17. Tsai, S. Q. et al. Dimeric CRISPR RNA-guided FokI nucleases for highly specific genome editing. Nat. Biotechnol. (2014).
18. Carroll, D. Staying on target with CRISPR-Cas. Nat. Biotechnol. 31, 807–9 (2013).
19. Bae, S., Kweon, J., Kim, H. S. & Kim, J.-S. Microhomology-based choice of Cas9 nuclease target sites. Nat. Methods 11, 705–6 (2014).
20. J. G. Doench, E. Hartenian, D. B. Graham, Z. Tothova, M. Hegde, I. Smith, M. Sullender, B. L. Ebert, R. J. Xavier, D. E. Root, Nat. Biotechnol., 2014, DOI:10.1038/nbt.3026.
21. H. Xu, T. Xiao, C.-H. Chen, W. Li, C. Meyer, Q. Wu, D. Wu, L. Cong, F. Zhang, J. S. Liu, M. Brown, S. X. Liu, Genome Res., 2015, DOI:10.1101/gr.191452.115.

Boutros lab, E-CRISP-Version 5.4
For suggestions please contact us at crispr@dkfz.de