search

pinetree1 / crispr-dav

CRISPR NGS data analysis and visualization pipeline
MIT License
15 stars 15 forks source link

two target sequences for a single gene #2

Closed Young5198 closed 5 years ago

Young5198 commented 6 years ago

I am trying to mutant a gene with two target sequences. From my preliminary data, I found there some mutations induced by CRISPR/Cas9 have indels at both target sites, but this software can only produce editing events for each target sequence respectively. Are there any possibility that I can separate amplicons with mutations at two target sites from the amplicons with only target site mutation? Thank you.

pinetree1 commented 6 years ago

At the moment, the tool analyzes target sequences separately if there are two or more targets per sample. Analysis of them together is under development but not ready yet. I would suggest you load the bam and bed files into IGV to view the indels at the two targets. Thanks.

Young5198 commented 6 years ago

Thank you for your reply. Could I use the whole region between two target sequences as one target sequence to do the analysis? I tried that, but only the allele with the deletion of that whole region was count as a mutation. All the other reads including wt, indels at each target site cannot be picked up. Any thoughts?

pinetree1 commented 6 years ago

Let's say there are guides A and C, region B is between A and C. Since you are using B as guide, you would get indel reads that have indel inside B which potentially extend to A or C. However, if the indels occur to the left A or the right of C, then using B won't likely find those. So we have to look at a read with regard to A and C, not B. Furthermore, if there are 3 or more guides, your approach won't work either.

pinetree1 commented 6 years ago

A rough way to estimate the percentage of reads that have both guides would be to look at the .var file in the 'align' directory. The file lists indel depths at each position. You can manually calculate the indel rates for each guide. Keep in mind not all positions are on the same read. So the calculation is not exact, but could be a decent estimation.

Young5198 commented 6 years ago

Thank you.

Yang

On Mon, Jun 11, 2018 at 10:24 AM, Xuning Wang notifications@github.com wrote:

A rough way to estimate the percentage of reads that have both guides would be to look at the .var file in the 'align' directory. The file lists indel depths at each position. You can manually calculate the indel rates for each guide. Keep in mind not all positions are on the same read. So the calculation is not exact, but could be a decent estimation.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/pinetree1/crispr-dav/issues/2#issuecomment-396282882, or mute the thread https://github.com/notifications/unsubscribe-auth/AmJc_My9okntZRTSfz5GAkWBduFwe8QIks5t7ouQgaJpZM4UdAPZ .

-- Yang Liu Ph.D. Candidate Interdepartmental Plant Biology Department of Horticulture Iowa State University Ames, IA 50011

pinetree1 commented 5 years ago

The development for this functionality is delayed. If you run the pipeline twice, each time with a different guide for the same sample, then you can compare the files align/..tgt and align/..tgt of the 2 runs. Extract those entries that have the same read in both files. This would be the total reads spanning those 2 guides. Find the WT and indel entries in them to calculate percent indel.