title          :: Start to Finish Hands-On (1/17): Intro
presentation   :: afni16_start_to_finish.pdf
scripts        :: shell_commands.txt
date           :: 2020-04
duration       :: 15:40
speaker        :: Reynolds
outline        ::
Plan to use afni_proc.py for single subject analysis - understanding the method
+ it will write tcsh processing script (and execute it)
Mention uber_subject.py: needs updates to match those for afni_proc.py
+ multi-echo, physiological data, distortion correction, NL-registration, etc.
Want to run a group analysis, how do we get there?
+   -- so need standard space subject betas (effect size maps)
+   -- aligned to some template (probably choose based on preferred atlas)
+   -- so want processing script(s), single subject pre-proc through regression
+      inputs:   epi, anat, stim timing
+      outputs:  beta weights (effect size estimations), and a lot more, QC
+      controls: processing choices (template, voxel size, blur size, etc.)
script EVERYTHING
+ record of methods applied
+ repeat per subject
+ be prepared to re-analyze all data
+   - e.g., now all subject data is acquired, use single version of AFNI
+ DO NOT JUST PRESS SOME BUTTONS TO RUN ANALYSIS
Publish afni_proc.py command along with AFNI version used (e.g. AFNI_20.0.00)
Focus on methods, feel responsible
NO SINGLE CORRECT WAY TO PROCESS DATA
depends on acquisition, task, subjects, group analysis plans, etc.
Practice good habit of viewing data