Changes to Conn to allow saving of 4D preprocessed data, and small bu…

[rcwelsh]

@dankessler @dankessler @sripada

Allowing the code to preprocess the data, but not run correlations and allows the 4D data to be saved for later preprocessing. This can be a time saver for looking at the same preprocessed 4D data and using different ROI schemes or bandpass filtering etc.

Changes are:

SOM_WriteNII.m changed how file_array is called, directly pass parameters default to using vox_offset of 0.

ConnTool_batch_mc_central.m put in logic to handle processing flag ‘preprocsave’, removed ‘som’ switch as it’s not documented. instead of using “diff”, use “gradient” to calculate the derivatives.

ConnTool_batch_mc_template.m introduced ‘preprocsave’ as processing switch to preprocess and then save the 4D nifti file to the Output area.

Code/SOM_CheckDataStructure.m fixed bug that if confound times-series file was not specified it would not correctly default to using the signal time-series file, this threw an error. now set cP to be data.run.P properly and change order of checking so that P is checked fully and then cP is checked.

SOM_CalculateCorrelationMaps.m automatically calculate the z values from r with SOM_Rho2Z

I've tested locally on my mac with a single subject and passed the test. I'll need a +1 and somebody to merge into dev.

[rcwelsh]

@dankessler @mangstad @sripada This pull request is still dangling..........

[rcwelsh]

@dankessler @mangstad @sripada ping ping ping pong ping pong pong ping....