Is there a benchmark analysis for the best tool for multiome single-cell RNA+ATAC data analysis?

Question

[Dan]

Hello:

Which is best practice and tool for multiome single-cell data analysis? Which package is the best:

SnapATAC2 (https://kzhang.org/SnapATAC2/)
MuData (https://muon.readthedocs.io/en/latest/notebooks/quickstart_mudata.html)
Seurat (https://satijalab.org/seurat/articles/multimodal_vignette.html)
Is there a benchmark analysis for the best tool for multiome single-cell RNA+ATAC data analysis?

Thanks a lot!

[jared.andrews07]

Define "best" for your use case.

[Dan]

maybe which is more popular ?

[jared.andrews07]

That is a poor metric to define "best" for your use case.

I'd recommend reading a few reviews, best practices books, and benchmarking studies (which can often be found in review articles/googling).

[ATpoint]

Since Op is new to this I recommend to read the first few pages of every manual and go for the tool that you feel comfortable with. In typical analysis it all comes down to making the right decisions in terms of which steps and functions to use to answer a given question. You can only make this decision if you understand basic principles of what the tools and functions do, and for this you need good documentation. Hence, go for the tool with best documentation. Usually these heavily-used tools are not terrible and mostly do similar things so it really comes down to how well they've documented.

Answer 1

[judhenaosa]

I recommend the single-cell best practices across modalities paper, including the book with Python notebooks:

Paper:

https://www.nature.com/articles/s41576-023-00586-w#Sec18

Book:

https://www.sc-best-practices.org/preamble.html

Regards,

Juan