Hi Ricardo
Both family data and longitudinal data are special cases of clustered data. In longitudinal studies the clusters are composed of the repeated measurements obtained from a single individual at different occasions. Longitudinal data have the special characteristic that they also have a temporal order (the first measurement within a cluster necessarily comes before the second measurement, and so on) and this has important implications for the analysis. In family data, the clusters are composed of measurements obtained from the individuals of the same family. In both longitudinal and family data observations within a cluster will typically exhibit positive correlation.
In order to apply lme to your family data you must order your design matrix and MRI data (eg. cortical thickness) in a way that you have all the assessments for the first cluster (family) then all the assesments for the second and so on. Then you need to built a vector "ni" of length equals to the number of clusters in your data. Each entry of that vector contains the number of elements in the corresponding cluster (the vector is ordered according to the previous ordering of your clusters).
Alternatively you could use something like
[M,Y,ni] = sortData(M,1,Y,sID); (sorts the data)
as explained in the wiki. All you need to do is assign the same subject ID (sID) to the elements within a family and add a fake time variable to the matrix of covariates M. After sorting the design matrix an data you then remove the fake time variable from your design matrix.
Best regards -Jorge
De: Aldo Cordova elabore@hotmail.com Para: "freesurfer@nmr.mgh.harvard.edu" freesurfer@nmr.mgh.harvard.edu Enviado: Jueves 26 de septiembre de 2013 5:39 Asunto: [Freesurfer] Mixed effects: is data ordering always required?
FreeSurfer's LME wiki site states:
"For computational efficiency reasons, these tools require the data
ordered according to time for each individual (that is, your design matrix needs to have all the repeated assessments for the first subject, then all for the second and so on)."
The dataset I need to
process includes pairs (and trios) of brothers, and although they're definitely clustered statistical units, there is no pre-defined sorting of individuals within a family (at least in this study). Please note that mine is not a longitudinal study, and the need for a cluster-correction arises from brotherhood...
Is FreeSurfer's LME
still the appropriate choice? How should this dataset be processed in FreeSurfer for cortical thickness analysis? Would you recommend another option?
Regards, Ricardo _______________________________________________ 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.