Referee 2's Review (José-Enrique O'Connor):

The manuscript by Rosado et al. describes a novel dual fluorescence, live-cell based platform (BaFiso) designed to identify and screen selective inhibitors of Akt signaling. BaFiso is composed of a co-culture of isogenic Ba/F3 cells, that have been engineered to survive independent of IL-3 by different mechanisms and are individually tagged with either Enhanced Yellow Fluorescent Protein (EYFP) or Enhanced Cyan Fluorescent Protein (ECFP). Thus, the first cell line (named BYA) is EYFP-tagged and is transduced with myristoilated Akt (myrAkt) which constitutively inactivates pro-apoptotic downstream targets. The second cell line (named BCS) is tagged with ECFP and expresses STAT5A1*6, an activating mutation of STAT5 transcription factor. Thus, contrarily to the parental Ba/F3 cell line (an IL-3-dependent pro-B lymphoblastic cell line), in BYA cells, survival in the absence of IL-3 depends on the expression of activated Akt, while BCS cells survive by Akt-independent mechanisms. By means of automated high-content screeening (HCS) bioimaging instrumentation and appropriate algorithms for single-cell identification and quantification, the change in the relative cell number of the two lines can be measured and used to identify molecules that specifically interact with each of the signaling pathways.

Authors report here the procedures to construct and maintain BYA and BCS cell lines and prove the modifications introduced and their functionality in the context of Il-3 independent survival. Using robotic systems for cell culturing and liquid handling and a HCA bioimaging system authors show the suitability of this cellular system to high-troughput screening (HTS). Furthermore, by applying a battery of commercially available general cytotoxicants and selective inhibitors authors obtain a proof of principle, sustained by appropriate statistical analysis, of the suitability of the BaFiso platform as a live-cell system to identify specific inhibitors of Akt signaling. Thus, authors conclude that small molecules can then be screened in the BaFiso platform to identify inhibitors that rescue IL-3 dependence.

This assay represents an improvement over the previous cell-based assays and screens since it allows to treat in the same well both the target and the control cells, thus allowing to differentiate easily between general cytotoxicity or non-selective effects from specific responses mediated by the pathway selectively studied. This is a novel and relevant report in the field of cell-based drug screens, which would be of interest to the area of drug discovery, both in industry and academia, as well as to those interested in methodological developments in cell-based assays. This cellular assay could be easily adapted to multiparametric flow cytometry, a single-cell based technology that generates data in a faster way than HCA bioimaging.

The manuscript is clearly written. However, there are some minor style questions that should be addressed by the authors to improve the manuscript, as follows:
- The abbreviations of Enhanced Yellow and Cyan Fluorescent Proteins should be consistent (some times they are abbreviated as EYFP or ECFP, while in other cases they are just termed YFP and CFP).
- In the Conclusions/Significance part of the Abstract authors state that "Simple analysis of the fluorescence emitted by these paired cell lines in co-culture permits selective inhibitors of Akt signaling to be identified.". This is a misleading sentence, since the endpoint determined for a given test compound is not a simple measurement of fluorescence emission, but a calculation of the relative abundance of two distinct fluorescent cell populations, based upon rather complex algorithmic approaches applied to a non-trivial capture of fluorescent bioimages. Possibly, authors may want to underline that once having the BaFiso cell system exposed to a particular treatment, no complex manipulation of the samples (i.e. of the 96-well plates) is required to obtain data.
- In the second paragraph of the Introduction, authors state that "Each of these isogenic lines was intrinsically labeled with a fluorescent marker...". This sentence is misleading since among users of image and flow cytometry, the concept of labelling with fluorescent markers is largely understood as the process of staining cells with exogenous chemical or biological fluorochromes. I would rather suggest using the concept of fluorescent reporting.
- In the same sentence, authors state that "..and thus, the ratio of two spectrally distinct fluorescent proteins could be used as primary endpoint of the system to monitor pathway-specific cytotoxicity". Again, this may be misleading since, as commented above, the endpoint measured is not the ratio of fluorescent molecules (this is indeed a typical parameter in cellular fluorescence measurements) but the ratio of one cell population bearing a particular fluorescent reporter to another cell population expressing a different fluorescent protein.
- In a similar context, the statement appearing in the methods section "The ratios of the cyan and yellow fluorescence signals were determined by dividing the number of CFP positive cells by the number of YFP positive cells in each well..." is essentially incorrect and should be rephrased almost simetrically, i.e. "The ratios of the CFP positive cells to YFP positive cells were determined by dividing the number of cyan fluorescence-emitting single cells by the number of yellow fluorescence-emitting single cells in each well".

Authors might illustrate with references their initial statement in the Introduction section ("Cell-based screens have been widely used in drug discovery although historically, these assays are conducted using genetically diverse cell lines derived from human tumors...")

The manuscript provides detailed description of the cell-engineering procedures and the HTS settings and HCS of the BaFiso platform but more detailed description of the algorithms and data analysis required to generate meaningful data would be welcome and be required for on-line publishing of the BaFiso protocol. In addition, there are some minor questions in the Methods section:
- Authors should indicate how IC50 values of toxicity were calculated from cell viability raw measurements.
- The brand and/or location of some equipment should be indicated (e.g. multidrop automatic dispenser, BD Pathway bioimager...).
- The brand name Immobilion should be Immobilon.
- The acronym U.A.M. should be spelled.
- In the sentence and oxidized forms AlamarBlue...", the particle "of" is missing.

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.