Description

This track shows the in-silico design of crRNAs for Cas13 using the tool nCov2019_Guide_Design, as described in Abbott et al., 2020.

To target highly conserved regions of the SARS-CoV-2 genome, an in-silico collection of all 3,802 possible crRNAs were generated. After excluding crRNAs that are either predicted to have potential off-target binding (≤2 mismatches in the human transcriptome) or having poly-T sequences that may prevent crRNA expression (≥4 consecutive Ts), a set of 3,203 crRNAs were obtained. These crRNAs are also able to target SARS and MERS with ≤1 mismatch.

Each crRNA has been characterized with four features:

• Efficiency is predicted using the online tool at https://gitlab.com/sanjanalab/cas13
• Specificy is determined by the number of off-target loci in human mRNA with ≤2 mismatches to the crRNA
• Generality within Coronaviridae is quantified as the percentage of Coronaviridae strains targeted by the given crRNA with perfect identity
• Generality within SARS-CoV-2 is quantified as the percentage of 1,087 SARS-CoV-2 patient genomes downloaded on March 20, 2020 that are targeted by the given crRNA with perfect identity

Method

To design all possible crRNAs for the three pathogenic RNA viruses (SARS-CoV-2, SARS-CoV, and MERS-CoV), the reference genomes of SARS-CoV, MERS-CoV, along with SARS-CoV-2 genomes derived from 47 patients were first aligned by MAFFT using the --auto flag. crRNA candidates were identified by using a sliding window to extract all 22-nucleotide (nt) sequences with perfect identity among the SARS-CoV-2 genomes.

We annotated each crRNA candidate with the number of mismatches relative to the SARS-CoV and MERS-CoV genomes, as well as the GC content. 3,802 crRNA candidates were selected with perfect match against the 47 SARS-CoV-2 genomes and with ≤1 mismatch to SARS-CoV or MERS-CoV sequences. To characterize the specificity of 22-nt crRNAs, we ensured that each crRNA does not target any sequences in the human transcriptome.

We used Bowtie 1.2.2 to align crRNAs to the human transcriptome (hg38; including non-coding RNA) and removed crRNAs that mapped to the human transcriptome with ≤2 mismatches.

Data Access

The raw data can be explored interactively with the Table Browser, or combined with other datasets in the Data Integrator tool. For automated analysis, the genome annotation is stored in a bigBed file that can be downloaded from the download server.

Annotations can be converted from binary to ASCII text by our command-line tool bigBedToBed. Instructions for downloading this command can be found on our utilities page. The tool can also be used to obtain features within a given range without downloading the file, for example:

Please refer to our mailing list archives for questions, or our Data Access FAQ for more information.

Credits

The predictions for this track are produced by Xueqiu Lin and Augustine Chemparathy in Stanley Qi lab at Stanford University

References

Abbott, Timothy R., Girija Dhamdhere, Yanxia Liu, Xueqiu Lin, Laine Goudy, Leiping Zeng, Augustine Chemparathy, et al. , 2020. Development of CRISPR as a Prophylactic Strategy to Combat Novel Coronavirus and Influenza. bioRxiv