I have several hundred small peptide (10-20aa) sequences that came out of a screen. I want to get PDB files with predicted structures for all of them. I initially thought about Rosetta, but then learned that it doesn't work well with proteins that small. Furthermore, all of the aa sequences were randomly generated, thus they have no homology to existing proteins. I was wondering if an MD program such as NAMD would work for my purposes. If not, I'll probably just have to stick to secondary structure prediction.