SILAC data analysis

[HF]

Hi,

One of our collaborators has designed and performed multiple SILAC experiments. In this experiment, Light (L) and Heavy (H) labeled are two different biological conditions (L: WT and H: KO). After mixing L and H (1:1 ratio) and running the MS, we analyzed the data using MaxQuant (using Multiplicity 2). 

MaxQuant generates columns with H/L ratios. It also generates columns with separate L and H intensity. We usually analyze this data using Perseus by importing the normalized H/L ratios. 

Now my question is: can we also analyze the L and H intensities separately? in other words, I would like to consider the SILAC as a mix of two individual label-free samples. Is this possible?

I hope this question is clear. 

Thanks

HF

[Vladimir Gorshkov]

Hi, HF,

I can not see the reason why you can not. Technically, the first step of analysis of SILAC data is the same as for LFQ - feature detection. As a result, you have a list of LFQ abundances for all peptides in your sample, that are organized in SILAC pairs/triplicates, and, finally, relative quantitation is performed. The purpose of SILAC is to mitigate the possible difference in sample treatment, LC-MS performance, ionization etc, between the samples. All this becoming important if you want to use LFQ instead. That being said, I think you can treat SILAC experiment as a mixture of two LFQ ones, but I, honestly, do not understand why. My concern is that, MaxQuant might perform some "optimizations" (and there is no other confident way to know it, except asking the developers), if you indicate that the experiment is SILAC. For sure it will do identify both members of the pair/triplicate by a single identification (i.e. if one has heavy peptide identified, the identity will be transferred to the light one), but it might have some other "optimizations" as well (sorry, I am not really an expert in MaxQuant). Thus, I would be more careful to pick up the individual L and H intensities - they might not be calculated in the same way as in LFQ experiment.

Best regards,
Vladimir

[hzf...@gmail.com]

Thanks Vladimir for your thoughtful and quick response.