Can SPONGE be run with only six samples?

[HowardChen330]

Can SPONGE be run with only six samples?

thanks！

[mlist]

It would unfortunately not make much sense. SPONGE relies on partial correlation and in our experience ceRNA effects are moderate if not miniscule in many cases. To detect them you need a much larger number of samples. In our experience 50-100, depending on the data set.

[HowardChen330]

That’s unfortunate. I research a non-model fish species where public data is very scarce (at most ~10 samples). Including the six samples generated by our group, my total sample size is under 20. I found your package very interesting, so it’s a shame I can't apply it to my dataset. thanks！ ***@***.*** From: Markus List Date: 2025-12-20 06:56 To: daisybio/SPONGE CC: HowardChen330; Author Subject: Re: [daisybio/SPONGE] Can SPONGE be run with only six samples? (Issue #10) mlist left a comment (daisybio/SPONGE#10) It would unfortunately not make much sense. SPONGE relies on partial correlation and in our experience ceRNA effects are moderate if not miniscule in many cases. To detect them you need a much larger number of samples. In our experience 50-100, depending on the data set. ― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>