Hello Arda Balci

I would use LFQ values.

Did you inject your two samples multiple times? Ideally in triplicate (n=3) as a minimum as then you can perform statistical analyses in Perseus such as a Welch test.

Also, you could plot the data as a volcano plot and display the bait protein of interest and its interactors.

Let me know if you need any more help.

Good luck!

Best, Peter

Hi Peter Allen Faull ,

Thank you very much for your answer! It is very helpful.

I injected my triplicate samples at the same time (sequentially), so I can perform statistical analysis on them. I use volcano plot, however, when I use LFQ values my bait protein does not come up as significant (even if I play with FDR or S0 values). However, when I use iBAQ values a known interactor shows up as significant in the volcano plot despite of bait being non-significant again.

Is there a specific reason why would you use LFQ instead of iBAQ values to analyse samples ?

Also, do you try to analyse the data through excel ? For example, I do extract the LFQ values from ProteinGroups.txt and do either TTEST or ratio etc on them to see if there is any protein showing up significant or it shows up more in my sample compared to control etc. Do you think this is a way of analysis or should I just stick to Perseus ?

Thanks!

Arda

Hi Arda Balci

Yes, we have experienced the same issue with iBAQ / LFQ data.

In MaxQuant, there is a tab called "Group-specific parameters".

If you click on "Label-free quantification" and select "LFQ" from the drop-down menu, you could amend the "LFQ min. ratio count value" to 1 (from 2).

You'll then obtain more similar results in LFQ as you do in iBAQ (and your bait should return to being significant).

We use iBAQ and LFQ values frequently in the lab for different application. It all depends on the question being asked. And we do not use Excel for analyses - only Perseus as it allows you to perform all necessary statistical analyses and provides clear output plots.

Best,

Peter

Hi Peter Allen Faull ,

Again, thank you very much!

I have been trying to understand LFQ and iBAQ values, as well as learning MaxQuant and Perseus softwares, and this informations are incredibly useful. Very much appreciated.

One last question if I may ask; can you give me an example of an experimental setup that you would use iBAQ values over LFQ (or vice versa)?

In my case, I am simply trying to uncover the interacting partners of a protein (bait), therefore I am looking to see differentially abundant hits in my sample.

Kind regards

Arda