" ... we demonstrated that the entropy of the chemiluminescence time series is similar to random Poisson signals."
The problem that I see with this kind of experiments (and because that this section may be partially flawed) ss that don't take into account that:
1- As mentioned in  active biophotons (for example those glutamate induced) are those that are more biologically significant (information carriers) but there is a background biophoton emission that is more massive and is what is detected using solely photo counting techniques.
2- As is mentioned in various papers [2-5] the relevant biophotons are probably guided by biological waveguides (like microtubules or actin filaments) and are not those that are more detected externally.
3- There is incoming evidence that other factors that are hidden in a solely photocount analysis are information carriers, like spectral power densities, spectral analysis  and others, and that, in any case, biophotons carry information (see bystander and autooptic papers in subsection ).
And, as various papers point out in this website, reactive oxygen species are not the unique source of biophotons.
 Traill, Robert R. "Asbestos as' toxic short-circuit'optic-fibre for UV within the cell-net:—Likely roles and hazards for secret UV and IR metabolism." Journal of Physics: Conference Series. Vol. 329. No. 1. IOP Publishing, 2011.
 Tessaro, Lucas WE, Blake T. Dotta, and Michael A. Persinger. "Bacterial biophotons as non‐local information carriers: Species‐specific spectral characteristics of a stress response." MicrobiologyOpen (2018): e761.
Last modified on 04-Dec-18