It must be something in the difference of methods between the two studies. As linked prior (why I responded to my own post), brilacidin is in the 12,000 molecule database on reframedb.org, so it was tested, but didn't pass their screen for some reason.
I don't know which difference in methods is responsible, but don't think the lack of positive brilacidin results from the Nature study has to do with the proposed MOA. I think in Nature study they let the drug sit in culture for 16 hours prior to infection, so that could be one important difference, compared to 1 hour before or after infection for the RBL study. Also the amount of virus used to infect the cells was likely different, and that could affect the outcome. So many variables ...