Entering edit mode
2.0 years ago
yanbinwan
•
0
The current CRISPR sgRNA library targets the whole genome, but some genes in the cells used in the experiment are not expressed or have little expression. Is it effective to knock out these genes? If it is determined whether it is included in the scope of library screening according to the amount of gene expression, how to determine the amount of gene expression?
Is this an assignment? In any case it's out of context and lacks details.
We previously performed a genome-wide CRISPR screen in 293T cells, but generated a large amount of invalid data, so we wanted to redesign the library incorporating transcriptome-level data to reduce the workload of subsequent data analysis. By performing RNA-seq on wild-type 293T cells, we found that about 4000 genes had a calculated TPM value of 0, so it is believed that CRISPR Knockout targeting these genes may not have an effect. In order to narrow down the library, a screening of the targeted genes is required. The current challenge is to determine a limit so that genes whose expression levels are higher than this limit are included in our library.
I see. So what exactly is the question now, do you need a TPM cutoff? Wouldn't also the non-expressed genes be a valuable control to estimate the background levels of your screen? You could include this information into the analysis, like with MAGeCK, you could use the non-expressed genes as negative control genes so the permutation statistics can "learn" from it.
Thank you very much for your reply. In fact, we have only recently started to think about this question, and inspired by the article "CRISPR activation screen identifies BCL-2 protein and B3GNT2 as drivers of cancer anti-T cell-mediated cytotoxicity", we believe that knocking out low-expressing genes does not allow us to study their function , it was not sufficient as a control to achieve our goal, so we considered using CRISPRa for screening. We believe that for genes that are not expressed or underexpressed, even if knocked out, the effect on the cell is limited, and likewise, CRISPRa for highly expressed genes has an equally limited effect, so we wanted a TPM cutoff. Based on its value, we can split the genome-scale screen into knockout and CRISPRa.