Heeeeeeeeeeeeeeeeeeeeeeeeelp!

I have four different functional sets of data and two anatomicals in each subject. First two are aligned with first anatomical and last two are aligned with second anatomical. I also get the tlrc transformation in the same transformation.

@Align_Centers -no_cp -base /usr/local/bin/TT_N27+tlrc -dset anat1+orig -child SW1pre+orig SW2sw+orig
@Align_Centers -no_cp -base /usr/local/bin/TT_N27+tlrc -dset anat2+orig -child SW3post+orig SW4seed+orig

@auto_tlrc -suffix none -base TT_N27+tlrc -input ${ANA1}+orig
@auto_tlrc -suffix none -base TT_N27+tlrc -input ${ANA2}+orig

align_epi_anat.py -anat ${ANA1}_ns+orig -epi ${EPI2}_icor+orig \
-epi2anat \
-anat_has_skull no \
-epi_base 10 \
-child_epi ${EPI1}_icor+orig \
-volreg_method 3dWarpDrive\
-tlrc_apar ${ANA1}+tlrc

align_epi_anat.py -anat ${ANA2}_ns+orig -epi ${EPI4}_icor+orig \
-epi2anat \
-anat_has_skull no \
-epi_base 10 \
-child_epi ${EPI3}_icor+orig \
-volreg_method 3dWarpDrive\
-tlrc_apar ${ANA2}+tlrc

Following matrices are saved and available for later use
:
${EPI2}_icor_al_mat.aff12.1D
${EPI2}_icor_al_reg_mat.aff12.1D
${EPI2}_icor_al_tlrc_mat.aff12.1D

I want to first analyze data in original space and I have my statistical analysis maps. Then I want to transform them to tlrc space for later group analysis. When I tried to use 3dAllineate with ${EPI2}_icor_al_tlrc_mat.aff12.1D matrix to transform results into tlrc it didn't work. (I think my data is aligned and just needs tlrc functional transformation matrix which is not provided by align_epi_anat.py).

Now what is the best way to do things short of reanalyzing the data in tlrc space:

1)Using adwarp?
2)Using 3dAllineatte?
3)@auto_tlrc?
4) what base or master makes more sense to use with above programs: anatomical+tlrc or my functional+tlrc to transform functional result map to tlrc space?