Rerunning Tedana with RICA output

Hello!

I have a question about re-running tedana (ica-recalssify) after running Rica.

It says to include the manual_classification.tsv with the --ctab option, but it seems like that option is not present in the current tedana document.

Is it still possible to re-run tedana with this option?

(Apologies for the previous post! It posted an old draft of mine…)

Thank you!

Hi!

Thanks for using tedana and Rica!

I don’t know where you read that you have to pass manual_classification.tsv to ica-reclassify with the --ctab option. If I remember correctly, that option was available with the tedana command so you could rerun the denoising with your manual classification.

With ica-reclassify we have made it easier to reclassify components based on classifications made elsewhere (like Rica). You will want to use the accepted.txt and rejected.txt files that Rica gives you. So your command will look something like this: ica-reclassify --manacc accepted.txt --manrej rejected.txt and so on.

Hope this helps!

Hello,

Thank you for the quick response! I actually read it on neurostars, but the post was about 2 years ago so I just wanted to double check.

That makes sense! Although, whenever I downloaded the tsv file by clicking on save (from https://rica-fmri.netlify.app), it seemed like it gave me a single file named as manual_classification.tsv. Would I need to create my own accepted and rejected txt files? Or should I install RICA on my local device?

Thank you again!

Okay, that’s interesting. It is supposed to give you 3 files: accepted.txt, rejected.txt, manual_classification.tsv.

What version of Rica did you do your manual classification with?

Hello,

I am unsure which version it would be… I followed GitHub’s Rica link and assumed it was the most latest. Is there a way to know the version from the online version? Maybe I do need to install it on my local device!

Oh, there’s no need to install anything!

You can check the version by clicking on the About button.

If you don’t mind, could you please tell me what browser you are using so I can debug the issue?

I just made a quick test with the live version (v1.1.2) using Chrome and got all three files I mentioned above.

I see!

I have just checked and I am also on v1.1.2 as well, but I am using safari at the moment.

I will also give it a try it through Chrome and let you know if I get something different!

It works with Chrome! I have just received 3 files by clicking on save.

1 Like

Perfect! I think we will have to update the documentation to advice not using Safari.

1 Like