The simulations testing the optimal scheme presented assume a low 30% asymptomatic
fraction. The reduction in infected individuals present when using the strategy presented,
in comparison to pure random testing, increases strongly with the asymptomatic fraction
of the population in question. Although the exact effect of the strategy depends on the details,
for large values of this parameter, reductions in infected persons present in comparison to
pure random testing can grow up to a factor of 3, at constant number of daily tests. For example,
in the case shown in figure 3b, where the reduction was of a factor of 2 for the assumed 30%
asymptomatic fraction presented, assuming an asymptomatic fraction of 80% yields a factor of 3
reduction in infected persons present, in comparison to pure random testing, at constant number of daily tests.

We had been expecting to introduce the above comment on receiving proofs, when we were surprised
by the news that there are no proofs in PLOS ONE.