Hi Dan,

Actually, while I agree with your interpretation, I
also agree with how the script was written. The
problem is with your afni_proc.py options.

A good place to start is following the examples from
the afni_proc.py help output. I expect that every
example using the 'align' processing block includes
the -volreg_align_e2a option, but none of them include
'-align_opts_aea -epi2anat '.

So you should expect not to include it, unless you
have a good idea about how things work. In this case,
it is not appropriate.


The basic purpose of the -BLOCK_opts_PROGRAM options
(such as -align_opts_a2e) is to allow the user to add
options directly to the given program, without afni_proc.py
needing extra functionality. But that means afni_proc.py
has no idea what is being attempted, and it has no ability
to verify the usage.

In this case, you are reversing the effect of the align block
without afni_proc.py knowing about it. Instead, leave the
align block as it is (doing anat to EPI), and let afni_proc.py
properly invert it with the -volreg_align_e2a option.

- rick