I have 900 peptide sequences (10-20aa each) for which I'd like to get some info about their predicted interaction with the cell membrane. The sequences were all randomly generated (and then selected during a screen), so they have minimal homology to any aa sequences in nature. Any programs that are heavily dependent on homology-based prediction won't work well. I'm also looking for a program that I can download (i.e. not a webserver) because I have so many sequences. What would you recommend?

Thanks