Splatter: simulating fold changes for specific genes?
2
1
Entering edit mode
Brian Haas ▴ 90
@brian-haas-6657
Last seen 3.0 years ago

Hi - I have an existing set of single cell data that I can use to have Splatter learn the initial set of parameters.  Then, I'd like to adjust the relative gene expression of specific genes so that they'll be simulated at specific higher or lower target levels.  Is there a straightforward way to do this?

My naive thought was to just multiply the original count data by whatever fold change factor before it goes into the Splatter parameter estimation, but I'm sure this is suboptimal if not just simply inaccurate.

Many thanks!

~brian

splatter • 391 views
ADD COMMENT
2
Entering edit mode
luke.zappia ▴ 50
@lukezappia-11973
Last seen 7 months ago

Hi Brian

Thanks for giving Splatter a go. I'm going to assume you are using the Splat simulation but you might want to check out some of the others depending on your use case.

The Splat estimation process fits distributions to the data and there is no direct connection between an individual gene in your real dataset and an individual gene in the simulated dataset so modifying your original data probably won't have the effect you want.

All the DE factors are sampled from a distribution which you can control using the related parameters. If you want to alter them for specific genes my suggestion would be to manually step through the Splat simulation functions and manually overwrite those at the correct point. If you want to give that a go I can explain more about how to do that.

Hope that helps

Luke

ADD COMMENT
1
Entering edit mode
Brian Haas ▴ 90
@brian-haas-6657
Last seen 3.0 years ago

Thanks, Luke!   I've been going through the splatter code and exploring my options here.  The coding and documentation in splatter is absolutely spectacular!  and that's been a huge help.   I'll be in touch with questions.

ADD COMMENT

Login before adding your answer.

Traffic: 439 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6