homework 3 submission#12
homework 3 submission#12monicaycli wants to merge 19 commits intobircibrain:masterfrom monicaycli:master
Conversation
fix -regress_censor_motion update directory in Exp-1.sh
also delete redundant output files
| # Part 2: Cluster correction | ||
| ## Estimate Data Smoothness | ||
| for f in blurs/blur_est.*.1D; do | ||
| 1d_tool.py -infile $f -select_rows 3 -write - >> blurs.acf.1D |
There was a problem hiding this comment.
Not quite. I gave you the REML results, which are on a different row.
| echo "* voxel-wise threshold = .001, cluster-wise threshold = .05 " >> Words.group_clusters.md | ||
| echo "* coordinates are peak voxel of each cluster in MNI space " >> Words.group_clusters.md | ||
| echo " " >> Words.group_clusters.md | ||
| echo "| Cluster Size | Peak Value (Z) | x (mm) | y (mm) | z (mm) |" >> Words.group_clusters.md |
There was a problem hiding this comment.
What are the cluster size units?
| @@ -0,0 +1,241 @@ | |||
| HISTORY_NOTE = [mli@Ringo-Lite.frontier.com: Fri Aug 18 19:14:42 2017] {AFNI_17.2.06:macosx_10.7_local} 3dttest++ -mask full_mask.group+tlrc -prefix Words.group -setA 'betas/stats*+tlrc.HEAD[Words#0_Coef]'\n[mli@Ringo-Lite.frontier.com: Fri Aug 18 19:29:06 2017] {AFNI_17.2.06:macosx_10.7_local} 3drefit -atrstring AFNI_CLUSTSIM_NN1_1sided file:cluster.NN1_1sided.niml -atrstring AFNI_CLUSTSIM_MASK file:cluster.mask -atrstring AFNI_CLUSTSIM_NN2_1sided file:cluster.NN2_1sided.niml -atrstring AFNI_CLUSTSIM_NN3_1sided file:cluster.NN3_1sided.niml -atrstring AFNI_CLUSTSIM_NN1_2sided file:cluster.NN1_2sided.niml -atrstring AFNI_CLUSTSIM_NN2_2sided file:cluster.NN2_2sided.niml -atrstring AFNI_CLUSTSIM_NN3_2sided file:cluster.NN3_2sided.niml -atrstring AFNI_CLUSTSIM_NN1_bisided file:cluster.NN1_bisided.niml -atrstring AFNI_CLUSTSIM_NN2_bisided file:cluster.NN2_bisided.niml -atrstring AFNI_CLUSTSIM_NN3_bisided file:cluster.NN3_bisided.niml Words.group+tlrc~ | |||
There was a problem hiding this comment.
Fun fact - Ringo means apple in Japanese and it also happens to be my favorite Japanese singer's name. Hence the name for my laptop.
| * q = .05 | ||
| * coordinates are center of mass of each cluster in MNI space | ||
|
|
||
| | Cluster Size | Peak Value (Z) | x (mm) | y (mm) | z (mm) | |
There was a problem hiding this comment.
Note that these are corrected z values. You would have to run 3dclust on the uncorrected z values, masked by the thresholded FDR-corrected Z map to get uncorrected Z values.
| -both \ | ||
| -pthr .05 .01 .001 \ | ||
| -athr .1 .05 .025 .01 \ | ||
| -iter 2000 \ |
There was a problem hiding this comment.
This is ok for HW and in class, but use more iterations (at least 5000, preferrably the default of 10,000) for any real analysis.
|
hw3 now updated based on @rhancockn's comments! |
No description provided.