I have created new experiments uing the SGA procedure and put the inoc and image logs of version2 and version 3 of the cSGA experiment into ROD3, there was a problem with the memory running out on the machine and the upload of tha image log stopping randomly. We got round this by telling ROD (I think) to use more memory however whilst this has sorted the problem temporarily I think bigger data sets will be impossible unless inputted in small chunks.