Ok, I checked out with the people who work with the transformation of the data(transfer the original bvec and bval to the BIDS structure, i think she made some mistake when transferring it), Thanks very much:)

De : freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> de la part de Yendiki, Anastasia <AYENDIKI@mgh.harvard.edu>
Envoyé : mardi 3 janvier 2017 22:50
À : Freesurfer support list
Objet : Re: [Freesurfer] Error from trac-all prep: data and bvals/bvecs do not contain the same number of entries

Hi Hao - Every single image in your DWI series should have a corresponding column (or row) in the b-value and gradient table. So if the first image in your DWI series is a b=0 image, and it doesn't have a corresponding column (or row), you should add 0 to the b-value table and "0 0 0" to the gradient table.

Hope this helps,

a.y

From: freesurfer-bounces@nmr.mgh.harvard.edu [freesurfer-bounces@nmr.mgh.harvard.edu] on behalf of Hao wen [hao.freesurfer@hotmail.com]
Sent: Wednesday, December 21, 2016 12:18 PM
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] Error from trac-all prep: data and bvals/bvecs do not contain the same number of entries

Hello:

I followed the tutorial for tracula, because of the tutorial data that you give dont have the T1 image, so I use on of my own diffision data and the reconall-ed T1 image, and I always get the same errors:

data and bvals/bvecs do not contain the same number of entries ,

So I guessed that it is the problem from the data(dwi.nii, T1.nii, .bvec, bval), but I tried to figure out the data dimension differences between my data and the tutorial data(tutorial_data/diffusion_tutorial/elmo.2005/orig/656000-11-1.dcm), Here is the output:

1) FS tutorial data(dicom format):

This is the bval info and i guess the bvec dimension is the same

bValue = 700

nB0 = 10

nDir = 60

this is the dwi.dicom information: dimensions: 128 x 128 x 60 x 70

So the nFrames is equal to the bvel and bval columns(70), which makes sense.

2) My data information(which is .nii.gz format):

The bvec and bval info:

number of columns for bval and bvec is 136

The dwi.nii.gz dimensions: dimensions: 96 x 96 x 59 x 137

So for my data, the number of columns(or transpose the files, the same spirit) is nFrames-1, I guess that is the problem, but my data can be run for Mrtrix and Dipy, even in the dipy tutorial, the data that they offer, the number of columns is the number of frames(volumes) minus 1, So for TRACULA, is it a mandatory that the number of columns in bvec and bval is equal to the number of volumes in dwi image???

I know sometime people dont write the first 0 in bvec and bval, so if I just add it, and run it for TRACULA, it will work??? I will try it tomorrow!!!

Really need your help, I stuck for this in a while:) the log file is attached.

Thanks in advance