# Subject source estimates on fsaverage brain model

**URL:** https://mne.discourse.group/t/subject-source-estimates-on-fsaverage-brain-model/3847
**Category:** Support & Discussions
**Tags:** source-localization, eeg
**Created:** [October 21, 2021, 11:44pm UTC](https://mne.discourse.group/t/subject-source-estimates-on-fsaverage-brain-model/3847 "2021-10-21T23:44:18Z")
**Posts on this page:** 3
**Page:** 1

<div class="post-metadata">

### Author: ![manorama](https://avatars.discourse-cdn.com/v4/letter/m/91b2a8/32.png) [@manorama](https://mne.discourse.group/u/manorama)
#### Post date: [October 21, 2021, 11:44pm UTC](https://mne.discourse.group/t/subject-source-estimates-on-fsaverage-brain-model/3847/1 "2021-10-21T23:44:19Z")

</div>

- MNE-Python version: 0.23.0
- operating system: Linux Ubuntu

Hi!  
I am trying to plot the sources for event related EEG from our subjects. I used a forward model from fsaverage MRI to get a uniform forward model for all subjects.

The alignment of EEG and MRI coordinates looks good, but the inverse solution shows sources not covering the whole brain.

![alignment](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/a85aa00e840b7ebb837690c3652f267d765364c5.jpeg)

![after](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/b351a3ca48c566d6b21a2644ec54ed65e9b80fe1.png)

Earlier, I had used patient-specific forward model and the sources were distributed on all of the brain lobes.

![before](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/57f6eefa6abe8ee676622e5f9493899c11042987.png)

Does anyone have an idea what has gone wrong?

Thanks,

Manorama

---

<div class="post-metadata">

### Author: ![richard](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/richard/32/15_2.png) [@richard](https://mne.discourse.group/u/richard)
#### Post date: [October 22, 2021, 5:18am UTC](https://mne.discourse.group/t/subject-source-estimates-on-fsaverage-brain-model/3847/2 "2021-10-22T05:18:47Z")

</div>

Hello @manorama,

are you saying that you do have subject-specific MRI scans available? Then what you want to do is run the entire source modeling using those.

Only at the very end, you’ll want to morph the source estimates to fsaverage so you can average / compare results across subjects.

See the following example:

[https://mne.tools/stable/auto\_examples/inverse/morph\_surface\_stc.html](https://mne.tools/stable/auto_examples/inverse/morph_surface_stc.html)

Best wishes,  
Richard

---

<div class="post-metadata">

### Author: ![manorama](https://avatars.discourse-cdn.com/v4/letter/m/91b2a8/32.png) [@manorama](https://mne.discourse.group/u/manorama)
#### Post date: [October 22, 2021, 3:45pm UTC](https://mne.discourse.group/t/subject-source-estimates-on-fsaverage-brain-model/3847/3 "2021-10-22T15:45:14Z")

</div>

Hello @richard ,

Yes, I did try running entire modeling using subject-specific MRI at first, but the results looked unexpected and weird.

I wanted to see if the sources are being affected more by noise in the EEG or some fault in the forward model. So, we decided to try using forward model from the fsaverage template, and just use EEG from the subjects. Is this not possible?  
If not, then I will go back to the first approach.

Thanks!  
Manorama