Major concerns:
- The experimental design is very impressive, considering the large number of comparisons included in the experiment. A major concern for us was the seeming lack of biological replicates, or at least an obvious description of them. The methods section indicates that each infection experiment was carried out in a 100 mL jar, but it seems like there was one jar per treatment combination (it is possible we are missing something in our reading). We feel it is important to clearly state how many jars were prepared per treatment and if it was only one jar per treatment to discuss why this was the case. We understand that this is large experiment and it was likely very time-consuming to maintain all of the treatments, which could have made replication simply not feasible. However, biological replication would greatly increase the reliability of the results and it is important to recognize and discuss this. This comes especially into play when the authors discuss the variability in results between the current experiment and a previously published experiment. If there is biological replication in both experiments, readers can better evaluate whether the observed variation was within a normal range of variation for this system. Likewise, error bars in Figure 2 would make the results easier to evaluate. If we have interpreted the text properly and biological replication was not included, the authors could consider explaining why this was the case and explicitly how they account for it in the results.
- In the introduction, methods, and results section the authors refer to parasite "strains," but switch to parasite "isolates" in the discussion where it becomes clear that this distinction is rather important to interpreting the results of the study. When they are referred to as strains, we assumed that each strain was a genetically distinct and genetically homogenous culture. In the discussion, there is a more comprehensive description of the "strains" and they were more accurately described as "isolates." Although the isolates are more representative of natural infections, their very nature makes the experiments very difficult to replicate. It would be helpful to use more controlled isolates (known strain composition) and do 16S monitoring to determine if strain or isolate composition itself evolves and changes over time .