How to calculate protein fold changes in more than two groups

Question

[e.scollo]

Hi All

I am analysing a total of four different groups and my aim is to understand what are the proteins differentially expressed in the groups and their relative fold changes.

I am carrying LFQ with a bottom up approach.

I am aware that I can use ANOVA to assess what proteins are significantly different in between the groups, while if I want to know which proteins are specifically different in each group, a post-hoc test such as Tukey test should provide this information.

My questions is the following: If I wanted to express fold change for each protein within the four groups analysed, should I use the ratio of the group with highest response against the group with the lowest response for each protein, or would you use a different approach to report fold changes within the four groups? Just to give you an example I have the following table with responses for Protein A in the different groups:

Protein                 Group 1 Group 2 Group 3 Group 4 Mean    Fold change

 A                             100             400     600      100    300            6

The response for each group is the mean from three replicates. So in this case should I express the fold change as 6, or do you think I should use a different approach?

Thanks a lot for your help

Answer 1

[Kevin Blighe]

Fold changes are fold-changes - order in terms on nominator and denominator is important. When presenting these values, one must also explain the order of comparison such that the correct interpretation can be made.

In your case, I do not believe there is any right or wrong way to generate the fold changes. It heavily depends on how you want to explain the results to your audience and which groups are your focus.

For example, if group 2 was my main focus group, then I would say that: group 2's expression is 4 times that of group 1

If group 1 was my main focus group, I would say: group 1's expression is 1/4 that of group 2.

Kevin