[color=#000000]Hi [/color]
[color=#000000]In these scenarios (multiple users working on same source data on different CONN projects) I typically recommend using CONN's option to copy the raw data into local BIDS folders. You may do so using any of the two following procedures:[/color]
[color=#000000]A) if using the GUI to import your raw data (in CONN's [i]Setup.Functional[/i] or [i]Setup.Structural[/i] tabs) simply change the option that reads "import selected files" to "copy to local BIDS folder and import".[/color]
[color=#000000]or B) if using batch scripts to import your raw simply add a "batch.Setup.localcopy=true" field to your batch structure. [/color]
That will create one local copy of your raw data separately for each CONN project that uses that same data, so that all preprocessed files will be stored in those project-specific folders. If (e.g. for space-saving reasons) you prefer to avoid having multiple copies of the same raw data, one undocumented option to do so would be to edit in the file conn_importvol2bids.m the line that reads "[i]SOFTLINK=false[/i];" and change that to "[i]SOFTLINK=true[/i];". That will create soft links to your raw data files in the project-specific folders instead of actually copying the raw data when importing it (so that preprocessed files will still be stored in these folders but the original data does not need to be replicated too many times). This option has not been thoroughly tested (and it only works on linux or mac systems) so please let me know if you run into any issues
Hope this helps
Alfonso
[i]Originally posted by tkarsten:[/i][quote]Hello,
I'm wondering if there is any way to save the preprocessing files Conn outputs in a directory other than where the raw data came from? We have multiple users in our lab using the same dataset and running different analyses (and preprocessing pipelines). To prevent overwriting one another's data, we've been copying the raw data into our personal directories and running from there so the files outputted during the preprocessing steps aren't placed in the master directory, but that is starting to create a space issue. I've been perusing through the forum and haven't seen any posts related to my question. I've also been trying to look through the conn scripts to see if there was a fix that way, but that avenue hasn't been fruitful thus far.
Thanks in advance for any help/advice you can provide![/quote]
[color=#000000]In these scenarios (multiple users working on same source data on different CONN projects) I typically recommend using CONN's option to copy the raw data into local BIDS folders. You may do so using any of the two following procedures:[/color]
[color=#000000]A) if using the GUI to import your raw data (in CONN's [i]Setup.Functional[/i] or [i]Setup.Structural[/i] tabs) simply change the option that reads "import selected files" to "copy to local BIDS folder and import".[/color]
[color=#000000]or B) if using batch scripts to import your raw simply add a "batch.Setup.localcopy=true" field to your batch structure. [/color]
That will create one local copy of your raw data separately for each CONN project that uses that same data, so that all preprocessed files will be stored in those project-specific folders. If (e.g. for space-saving reasons) you prefer to avoid having multiple copies of the same raw data, one undocumented option to do so would be to edit in the file conn_importvol2bids.m the line that reads "[i]SOFTLINK=false[/i];" and change that to "[i]SOFTLINK=true[/i];". That will create soft links to your raw data files in the project-specific folders instead of actually copying the raw data when importing it (so that preprocessed files will still be stored in these folders but the original data does not need to be replicated too many times). This option has not been thoroughly tested (and it only works on linux or mac systems) so please let me know if you run into any issues
Hope this helps
Alfonso
[i]Originally posted by tkarsten:[/i][quote]Hello,
I'm wondering if there is any way to save the preprocessing files Conn outputs in a directory other than where the raw data came from? We have multiple users in our lab using the same dataset and running different analyses (and preprocessing pipelines). To prevent overwriting one another's data, we've been copying the raw data into our personal directories and running from there so the files outputted during the preprocessing steps aren't placed in the master directory, but that is starting to create a space issue. I've been perusing through the forum and haven't seen any posts related to my question. I've also been trying to look through the conn scripts to see if there was a fix that way, but that avenue hasn't been fruitful thus far.
Thanks in advance for any help/advice you can provide![/quote]