A macro-enabled Excel workbook (CRISPRn-annotator.xlsm), which could be used to quickly select paired-sgRNAs for CRISPR-mediated gene knockout experiments. A similar Excel workbook (CRISPRai-annotator.xlsm) can be used for picking sgRNAs for gene activation/inhibition experiments. All sgRNAs were collected from the previously reported genome-wide sgRNA libraries. Excel files automatically download the genbank files from NCBI, and annotate the CRISPRs.
Paired-sgRNAs co-transfected together with wildtype Cas9 into human/mouse cells would result in achieving a higher knockout efficiencies. Simultaneous DNA cuts at the targeted gene are repaired with a high precision and thus introduce a precise deletion (Ref. 1). Thanks to several genome-wide pooled CRISPR screening studies, a dozen of sgRNAs have been designed against each of the human and mouse genes. Therefore, these libraries offer a straightforward way for picking sgRNA-pairs for gene knockout experiments. These libraries have been designed and went under several rounds of optimization to particularly use the sgRNAs with optimal efficacy and minimal off-target activity. These libraries cover almost all coding and non-coding genes (e.g. microRNAs and lncRNAs). I have tested several sgRNA design tools and I often (not always) ended up with the same set of sgRNAs. So, I made my life easy with making a collection of these libraries (listed here: AddGene CRISPR Libraries) and prepared two Excel workbooks to curate the collection for a user-specified gene. Moreover, the sgRNA_Annotator workbook can predict the deletion size for different sgRNA pairs and annotate them on a genbank file (either provided by user or automatically downloaded from NCBI RefSeq).
-
Windows
-
MacOS
-
Microsoft Excel 2016 or higher
-
Internet Connection
- None, just enable macros using the pop up bar, which appears upon opening the workbooks.
-
CRISPRn-Annotator.xlsm: annotates CRISPRs used for knockout.
-
CRISPRai-Annotator.xlsm: annotates CRISPRs used for activation and inhibition.
- Download the library and the “sgRNA_Annotator_v1.0.xlsm” workbook from my GitHub repository.
- Unzip the downloaded files and make sure that all the .txt files exist in the "Library" folder:
- Open the “CRISPRn_Annotator_v1.0.xlsm” for gene knockout or “CRISPRai_Annotator_v1.0.xlsm” for gene activation/inhibition and click on the “Enable Content” button that appears on the yellow bar under the menu ribbon:
- Provide an official gene symbol, select the species (Human or Mouse), and click on the “Import” button:
- If you already have the genbank file for your targeted gene, click on the “Browse” button and select the file to be annotated. It is important to provide a gene sequence (and not a transcript). Alternatively, click on the “Download RefSeq” button, and wait for the Excel to download the genbank file. Annotated file can be found in the same folder that the Excel workbook is.
- Check the Paired-sgRNA combination table. Note the "Frame" column, some combination of the sgRNAs may results in inframe deletions, which is not ideal for the knockout purposes.
-
Adjust the distance between cut_sites if required and click on the “Update Paired-sgRNA” button for your changes to be applied.
-
Import the generated genbank file into your software of choice and enable the annotation types, the sgRNAs are annotated as “CRISPR” by default.
- Selected sgRNAs can be directly converted into oligo duplexes for cloning purposes. An example of the formula for the Fwd and Rev oligos are provided below:
- Finally, check the "Log" worksheet where errors are captured. Reset the workbook before closing the file.
-
Consider the RNA polymerase III and U6 promoter limitations if the sgRNA will be expressed using U6 promoter:
-
Avoid “T” homopolymers within the target site sequence.
-
Consider an additional “G” upstream of the target site (if the High Fidelity Cas9s will be used).
-
Some of the libraries have already taken these considerations into account, but not all.
-
Ensure sgRNAs target all the transcripts.
-
Ensure sgRNAs hit the first 50% of the CDS.
-
Ideally, there is no cryptic ATG after the target sites.
-
The genetic analysis method:
-
precise deletions after paired-sgRNA targeting would allow rapid detection of the editing efficiency on a simple agarose gel. However, deep targeted sequencing can provide a better insight into the resulting alleles. The amplicon size for deep targeting sequencing is usually short (less than 300 bp). Larger a deletion, a larger bias will be introduced in the amplification. Moreover, large deletions will require additional bioinformatics analysis of the NGS data.
-
My recommendation is to keep the deletion shorter than 100 bp.
-
Target sites with a “T” at 4th nucleotide before PAM tend to result in higher single nucleotide insertion Ref. 2 and in principle higher percentage of frame-shifting mutations.
CRISPRai_Annotator Excel workbook can be used for CRISPRa and CRISPRi applications. There are two additional inputs that need to be specified; Modulation mode (activation/inhibition) and the promoter length:
All CRISPRn/a/i sgRNAs can be annotated on the same genbank file:
-
The maximum size of a gene that can be downloaded is limited to 300,000 bp.
-
If the targeted gene is larger than 300kb, then only the first 300K will be downloaded.
Bonus: sgRNA Annotator can be used for more general tasks like downloading the RefSeq for a gene or annotating any other sequences on genbank files (e.g., primer sequences on a plasmid map, etc).
Please report issues and bugs to Amir.Taheri.Ghahfarokhi@gmail.com
Ref. 1 - Geisinger j., et al. (2016) In vivo blunt-end cloning through CRISPR/Cas9-facilitated non-homologous end-joining. Nucleic Acids Research, 44(8): e76.
Ref. 2 - Taheri-Ghahfarokhi A., et al. (2018) Decoding non-random mutational signatures at Cas9 targeted sites. Nucleic Acids Research, 46(16): 8417–8434.