How to correctly report exon number for CRISPR-Cas9 knockout when genes have multiple isoforms?

Question

[jakob.vanek]

Hi everyone,

I'm working on CRISPR-Cas9 knockouts of several SLC transporters in order to study their roles in Glioblastoma. Since these genes have multiple isoforms, I'm unsure what the best practice is when referring to the targeted exons in my thesis. So far, I've done the following: I used the longest isoform as a reference and specified both the exon number and the isoform name used in the annotation. For example: My question is: Is using the longest isoform and stating the exon number + isoform name an accepted and reproducible way of reporting CRISPR targets in such cases? Or is there a preferred convention in the community — e.g., referring to genomic coordinates, CCDS, or using transcript IDs (like RefSeq/Ensembl IDs) instead?

I want to make sure the data is understandable and reproducible for others, especially since exon numbers vary between isoforms.

Thanks a lot for your help!

[ATpoint]

I cannot say I have a reference, but genomic coordinates are simple, exact and easy to find back. I would report Ensembl gene ID and intended isoform, plus the exon number relative to that isoform, plus its genomic coordinates and the target sequence. That should do for reproduction. Maybe a screenshot from the genome viewer does further illustrate it, but the coordinates alone should actually do. Technically, since CRISPR acts on genome and not transcriptome it's probably the most correct way to report genome coordiates, after all.

Answer 1

[Kevin Blighe]

Yes, using the longest isoform as reference with exon number and isoform name is acceptable, but it's not the most robust for reproducibility due to varying annotations across databases and updates.

A better convention is to report the Ensembl or RefSeq gene ID, the specific transcript ID (e.g., ENST000003something for Ensembl), and the exon number relative to that transcript. This ties your targeting to a stable, versioned reference.

Additionally, include genomic coordinates (e.g., chr7:140453136-140453155, hg38 build) for the targeted site, as CRISPR acts on DNA, not RNA. This allows easy lookup regardless of isoform numbering changes.

Finally, provide the full gRNA sequence and PAM for exact replication. If possible, add a genome browser screenshot or link to illustrate.

This approach ensures clarity, especially for SLC transporters with multiple isoforms in glioblastoma studies.

Kevin