When we have only one sample (or maybe even just one replicate), we have a hard time estimating the log2 fold change distribution and the number of differentially expressed genes. In place of estimates, we can use a callback function to generate numbers as needed. The body of the function just needs to return numbers from a specified distribution with parameters you decide. Here, we'll build a function that returns numbers with a normal distribution of mean 0 and standard deviation 2. This reflects that we think the log fold change distribution is normal with these parameters. When we've built the function, it gets used in the DESetup() function in place of the vector of log2 fold changes. For the proportion of genes differentially expressed, we just have to guess or take an estimate from something we already know about the experimental system:

log2fc_func <- function(x){ rnorm(x, 0, 2)} 
prop_de = 0.1
de_opts <- DESetup(ngenes=1000,
 nsims=25,
 p.DE = prop_de,
 pLFC= log2fc_func,
 sim.seed = 58673
 )