Those statements above about gastric ischemia induced by exercise, even in healthy individuals, are true. Part of the "flight or fight" response shuts down digestion in favor of more urgent concerns. This is a non-trivial process controlled by the autonomic nervous system, and we have widespread indications many ME/CFS patients have some degree of dysautonomia, though we don't know the cause.
The idea of bacterial translocation in the gut causing symptoms of PEM seems promising. That malaise is regularly described as "flu like", and influenza definitely does cause similar disturbances in the gut. I would also point out that the time scale on which symptoms of a brief flu disappear is not too different.
Instead of trying to pick out useful information from the data published here I'm going to suggest a measurement which should be much simpler. If bacterial translocation is taking place more seriously when ME/CFS patients exercise, some common bacteria should be well represented. In particular e. coli., a proteobacterial sp., is likely to leave components in the blood, even if living bacteria do not reach the bloodstream and cause sepsis. There is a real connection here with possible mitochondrial dysfunction because the membrane of e. coli., like that of mitochondria, contains substantial cardiolipin.
I'm suggesting this because there have been inconsistent reports of anti-cardiolipin antibodies (ACA) in ME/CFS patients for years. If these are generated in response to bacterial translocation, and wash out over a number of days without exercise, the time scale might fit. Since we have no idea of the relation between previous patient exercise and past ACA measurements this could explain earlier inconsistent results.
Looking for a dynamic response to common pathogens strikes me as more likely to yield results than searching for rare pathogens when we don't even know precisely what we are looking for and what tissues to search.