Fig. 4.

Influence of the expression level of BD-Bait and AD-Prey on the reporter level. A–C, The expression levels of BD-Bait (1^st column), AD-Prey (2^nd) and reporter (3^rd) are visualized as a probability distribution of the hlog-transformed fluorescence intensities. In order to better visualize differences in the reporter expression we use an additional representation of the cumulative mean in linear scale (4^th column). The distributions and cumulative mean are shown for the global population in A and gated subpopulations in B and C. For reasons of clarity, only an illustrative subset of studied couples is shown. In the legend these couples are ordered and numbered (1-4) according to their in vitro affinity. The order of the couples according to the reporter expression (i.e. mean value of the Tag-BFP-H channel) is given on the very right side of the three subfigures. The global population of cells reveals significant differences in the expression levels of BD-Bait and to a smaller extent of AD-Prey A. Because of these expression level differences, the mean values of Tag-BFP-H are ordered differently than the in vitro affinities: 2 displays a stronger reporter expression than 1, and 4 stronger than 3. Only the discrimination between strong (1, 2) and medium (3, 4) interactors is possible. The successive gating of cells, first based on Tag-RFP-H (700 < Tag-RFP-H < 900) values (B), and then based on yEGFP-H (5000 < yEGFP-H < 6000) values (C), allows to equalize the expression levels. Such standardized subpopulations improve the extraction of quantitative information on the strength of the Bait-Prey interactions: the Tag-BFP-H mean value is ordered according to the in vitro affinities. The transparent red bar in the panel of the 1^st column shows the Tag-RFP-H gating interval applied to obtain the subpopulation in B (symbolized by the red action arrow). Similarly, the green bar and arrow illustrate the yEGFP-H gating interval applied to the subpopulation in B to obtain the final subpopulation in C.