There is an example of how to do this in the 454 docs, but it looks like the commands will just clobber the extracted single-read files (bcet.fasta, etc.) when extracting the paired-end files. Because of the input file naming scheme, it seems like there can only be one data set for each platform. Should I be doing one sff_extract with all the data at once (and just letting it try to find the non-existent linker in the unpaired data)? If it matters, I'm referring back to 3.0.0 docs, since the doc/docs/ folder seems to be missing in later releases... -Brian