Dear FS team, I would like to correct for multiple comparisons after I run command-line GLM analysis on stats tables. I would like to make a second level analysis between a control group and an autism group. I would like to know if there is a command-line in FreeSurfer or if there is any program or toolkit in Matlab that allows to do that.
Thank you, Laura
We don't have something explicitly to do this, but you can run FDR on it using either mri_fdr (use nomask for the mask) or you can load the sig.mgh file into matlab and run it there, eg,
sig = MRIread('sig.mgh'); p = 10^-abs(sig.vol); fdrthresh = fast_fdrthresh(p);
On 3/20/17 9:32 AM, Laura Ferrero Montes wrote:
Dear FS team,I would like to correct for multiple comparisons after I run command-line GLM analysis on stats tables. I would like to make a second level analysis between a control group and an autism group. I would like to know if there is a command-line in FreeSurfer or if there is any program or toolkit in Matlab that allows to do that.
Thank you, Laura
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear FS team, I would like to make an analysis between a Patients Group and a Control Group. This study is focused on identifying if there are differences in volume between both groups. I want to regress out the effects of gender, age and IQ.
My designed matrix has the following structure: GroupDescriptorFile1 Class PatientMale Class PatientFemale Class ControlMale Class ControlFemale Variables Age IQ Input subject1 PatientMale 25 110 Input subject2 ControlMale 30 90
I was thinking about a Contrast Matrix like this: 0.5 0.5 -0.5 -0.5 0 0 0 0 0 0 0 0 However, looking for some information I discovered that it only compares volume of two groups at the naissance because it is only comparing intercepts. Therefore, how I could make a comparison between volumes of two groups in general? Thank you, Laura
You should first test whether there is an interaction between group and covariate. If there is non, then you can keep the same FSGD file but run it with DOSS (different offset, same slope). The "same slope" means that the difference will be the same regardless of the value of the covariate. You will need to make a different contrast. See https://surfer.nmr.mgh.harvard.edu/fswiki/DodsDoss
for more info
On 4/16/17 7:04 AM, Laura Ferrero Montes wrote:
Dear FS team, I would like to make an analysis between a Patients Group and aControl Group. This study is focused on identifying if there are differences in volume between both groups. I want to regress out the effects of gender, age and IQ.
My designed matrix has the following structure: GroupDescriptorFile1 Class PatientMale Class PatientFemale Class ControlMale Class ControlFemale Variables Age IQ Input subject1 PatientMale 25 110 Input subject2 ControlMale 30 90
I was thinking about a Contrast Matrix like this: 0.5 0.5 -0.5 -0.5 0 0 0 0 0 0 0 0 However, looking for some information I discovered that it only compares volume of two groups at the naissance because it is only comparing intercepts. Therefore, how I could make a comparison between volumes of two groups in general? Thank you, Laura
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
In the case that exists an interaction between group and covariate. How a test comparing the mean volume of each group, like SPM does, could be done? I do not want to know if the different groups evolve in a different way with covariates, I only want to compare if groups have different volume in some regions of the brain. Or FreeSurfer does not allow to do a comparison like that? Thank you
Douglas Greve greve@nmr.mgh.harvard.edu escribió:
You should first test whether there is an interaction between group and covariate. If there is non, then you can keep the same FSGD file but run it with DOSS (different offset, same slope). The "same slope" means that the difference will be the same regardless of the value of the covariate. You will need to make a different contrast. See https://surfer.nmr.mgh.harvard.edu/fswiki/DodsDoss
for more info
On 4/16/17 7:04 AM, Laura Ferrero Montes wrote:
Dear FS team, I would like to make an analysis between a Patients Group and aControl Group. This study is focused on identifying if there are differences in volume between both groups. I want to regress out the effects of gender, age and IQ.
My designed matrix has the following structure: GroupDescriptorFile1 Class PatientMale Class PatientFemale Class ControlMale Class ControlFemale Variables Age IQ Input subject1 PatientMale 25 110 Input subject2 ControlMale 30 90
I was thinking about a Contrast Matrix like this: 0.5 0.5 -0.5 -0.5 0 0 0 0 0 0 0 0 However, looking for some information I discovered that it only compares volume of two groups at the naissance because it is only comparing intercepts. Therefore, how I could make a comparison between volumes of two groups in general? Thank you, Laura
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
You can do the comparison, but you should know what you are doing. If there is an interaction, then the question as to whether the two groups differ in mean does not make sense without specifying the value of the covariate where the difference is being measured. Just because a software package allows you to run an analysis does not mean it is a meaningful analysis to run. Having said that, 90% of the neuroimaging community would just do the test without appreciating the interpretation problems.
On 04/19/2017 12:43 PM, Laura Ferrero Montes wrote:
In the case that exists an interaction between group and covariate. How a test comparing the mean volume of each group, like SPM does, could be done? I do not want to know if the different groups evolve in a different way with covariates, I only want to compare if groups have different volume in some regions of the brain. Or FreeSurfer does not allow to do a comparison like that? Thank you
Douglas Greve greve@nmr.mgh.harvard.edu escribió:
You should first test whether there is an interaction between group and covariate. If there is non, then you can keep the same FSGD file but run it with DOSS (different offset, same slope). The "same slope" means that the difference will be the same regardless of the value of the covariate. You will need to make a different contrast. See https://surfer.nmr.mgh.harvard.edu/fswiki/DodsDoss
for more info
On 4/16/17 7:04 AM, Laura Ferrero Montes wrote:
Dear FS team, I would like to make an analysis between a Patients Group and aControl Group. This study is focused on identifying if there are differences in volume between both groups. I want to regress out the effects of gender, age and IQ.
My designed matrix has the following structure: GroupDescriptorFile1 Class PatientMale Class PatientFemale Class ControlMale Class ControlFemale Variables Age IQ Input subject1 PatientMale 25 110 Input subject2 ControlMale 30 90
I was thinking about a Contrast Matrix like this: 0.5 0.5 -0.5 -0.5 0 0 0 0 0 0 0 0 However, looking for some information I discovered that it only compares volume of two groups at the naissance because it is only comparing intercepts. Therefore, how I could make a comparison between volumes of two groups in general? Thank you, Laura
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear FS team,
Thank you very much for your help. Just a few more questions.
I want to perform a statistical analysis between two groups:controls and patients, using age and IQ as covariates. First I have to test if there is an interaction between group and covariate using DODS.
My questions are:
Q1) Do I have to test age and IQ separately, or together?
Q2) In case that there is not interaction in any of the covariates, can I use the same FSGD file but do I have tu run it as DOSS? How it could be done? I would appreciate if you could give me some hints or examples about the matrices and contrast files that I must use.
Thank you, All the best,
Laura
freesurfer@nmr.mgh.harvard.edu
-
Douglas Greve -
Douglas N Greve
-
Laura Ferrero Montes