search

Russel88 / CRISPRCasTyper

CCTyper: Automatic detection and subtyping of CRISPR-Cas operons
https://typer.crispr.dk
MIT License
89 stars 16 forks source link

Different results from CRISPR Cas typer web server and standalone version #34

Closed Smiriti-Gupta closed 1 year ago

Russel88 commented 1 year ago

I need a bit more information to figure out where the problem could be.

Smiriti-Gupta commented 1 year ago

Hello Sir, Thank You for the response. Yes, I am using the same standalone version as what is run on the web server (version 1.8.0). I am using this tool to identify CRISPR-Cas systems in Metagenome Assembled Genomes (MAGs).The tool identifies CRISPR-Cas systems in a particular bin through web server but no CRISPR-Cas are identified for the same bin using the standalone version. I also have certain other queries mentioned below:

  1. How can we use the standalone version to identify CRISPR-Cas system from all the bins at once. 2.Some of the Cas9 are predicted as ambiguous and the best type is shown as Type II-A and Type II-C, what does this mean? 3.The predicted Cas operon tab has a false positive column but then also it predicts the type and other details. Why?
  2. What does unknown gene actually mean (the gene is not known to play a role in CRISPR-Cas system or the function of the gene is not at all known)?
  3. One of the contigs shows two Cas9 (adjacent to each other). What does it signify?
  4. Cas1 gene is mostly shown by purple color in the graph plot, but it is a gene involved in adaptation so it should be shown in blue. What is the reason for this?

I will be greatful for a quick response. Thank You Smiriti Gupta

On Fri, 6 Jan 2023 at 14:18, Jakob Russel @.***> wrote:

I need a bit more information to figure out where the problem could be.

  • Are you using the same standalone version as what is run on the webserver?
  • In which way are the results different?
  • Do you have a reproducible example I can use for testing?

— Reply to this email directly, view it on GitHub https://github.com/Russel88/CRISPRCasTyper/issues/34#issuecomment-1373357602, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5CM5DJZDKSSV3BDYEFAVRDWQ7L6NANCNFSM6AAAAAATQZLEWI . You are receiving this because you authored the thread.Message ID: @.***>

Russel88 commented 1 year ago

Hi I will answer point by point. As for the difference between web server and standalone, how specifically are the outputs different? Do you use some non-default parameters on either the standalone or webserver?

1) You could put them all in one fasta and run cctyper with --prodigal meta, but it's faster to parallelize across MAGs with for example GNU parallel 2) There is a Cas9 HMM that matches both II-A and II-C, hence it's ambiguous 3) Not sure what you mean by a false positive column? 4) Unknown gene means that it is not CRISPR-Cas related 5) It's an operon with two cas9, or maybe it should have been one gene but the ORF was called wrongly 6) That sounds like a mistake, could you send one of the figures? Also, is it only for specific subtypes or for different subtypes?

Cheers, Jakob

Smiriti-Gupta commented 1 year ago

image Please find the image showing Cas1 in purple. It is for all subtypes

Russel88 commented 1 year ago

The Cas1 gene has the expected color. All adaptation related genes will have this color (blur, purple, or what you want to call it).