I am having some trouble finding a good definition of how to define a positive interaction (in relation to the negative and positive controls) when using a two-hybrid system to detect protein-protein interactions.
There are of course thousands of papers out there, but so far I've had no luck finding one that explicitly described how they decided on a cut out using similar assays as mine: I'm working with a system based on the bacterial two-hybrid (BACTH) system, and using an enzymatic assay to measure the activity of the reporter gene, ß-galactosidase.
I often hear about using a signal twice as strong as the negative control as the threshold. But what about the standard error of the negative control? Should I instead use twice the average + standard error as the threshold to be more stringent? Any other ideas?
I don't know the bacterial two-hybrid system. If possible, using more than one reporter gene is an approche to diminish the false positive like in the yeast two-hybrid system (Clonetech, four reporting genes).
Hi there,
What do you call a positive interaction in two-hybrid : the one giving a strong signal with the reporter gene? As Zhenlin said, try other screens and then at the end you will have to consider other approaches to validate the interaction you may have seen in two-hybrid.
In yest two hybrid the interaction is often identified by a selection assay on auxotrophic medium and by a b-galactosidase assay. Fixing the thresold above two time is probably a minimum, but it depends a little bit on the number of proteins identified.
If making an initial selection using a survival assay on auxotrophic mediun, the number of times a given interacting partner is identified can be indicative of the strenght of that interaction.
Some proteins produced often false positive (ribosomal proteins etc.). Maybe those can be put aside.
Once you have identified potential binding partner to your bait, you should make sure that the novel binding partner do not produce a positive signal in your assays when expressed alone or with a control construct lacking the bait fusion.
If your bait protein has a highly validated binding partner, it could represent a good control in your assays, particularly if point mutation disrupting the interaction are known.
In your case a positive interaction would be one that produces more ß-galactosidase than the bait+empty vector and prey+empty vector pairs. Rather than arbitrarily choosing 2-fold greater than empty vectors, you could perform a t-test between biological reps of negative controls and your bait+prey pair. A real interaction would be defined best as one that has statistically more ß-galactosidase than negative controls.
What is the range of the residual β-galactosidase synthesis that determines whether or not two proteins interact with each other?
On the basis of a series of results obtained with the two hybrid assay, you can assume that the residual level of β-galactosidase synthesis between X% and Y % represents positive interaction whereas values beyond this range are considered negative interactions. I know that this is an empirical determination. Data should be submitted to a statistical program, like SPSS, for cluster analysis and they will need to be confirmed by other methodologies.
Hi all! Sorry for taking so long to reply. First of all, thanks for your answers, it gave me a lot to think about.
Unfortunately, using different reporter genes isn't an option at the moment. This is a new two-hybrid system, developed in-house and we have yet to build new versions with new reporter genes.
In the end, I decided to simply go with a statistical approach, checking whether positive interactions are statistically different from the negative controls.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress