Feedback on anatomical delineations: disagreement on your mouse brain atlas

I would like to point out a potential mistake on your mouse brain atlas (P56, coronal) , regarding the labeling of visual cortical areas in your Mouse Brain Atlas (P56, Coronal).

Take, for example the image 83 of 132,

Here, what is labeled as VISam should be VISpm. In fact VISam is mislabeled all the way across the primary visual cortex (VISp) length in the rostro-caudal axis, as adjacent to VISp on the medial side. According to all studies detailing mouse visual cortical areas (e.g. Glickfeld and Olsen 2017; PMID: 28746815; Keller et al., Nature 2020, Figure 5 and Extended Data Figs 6-10; PMID: 32499655), it is impossible that VISam and VISp are adjacent in any coronal section. This is confirmed in your own Allen Brain Explorer (3d Viewer).

What you label as VISam is in fact VISpm, and in the reference coronal atlas, VISPm appears only at image 92 of 132, which is way too occipital.

Since this Atlas has become the gold standard to identify cortical areas, it will be a bit problematic for us to show virus infections aimed at VISpm using your coronal atlas…

Could you please look into it and possibly rectify it?

All the best


Thank you for raising this question!

We have two versions of adult mouse brain atlases, CCFv2 and CCFv3. The two atlases were annotated differently, especially in visual cortex including VISp. Our paper published in Cell this year detailed the difference between CCFv3 and CCFv2.

The problem you found exists in CCFv2 but in CCFv3. The Allen Brain Explorer (3D viewer) is generated from CCFv3 . The papers you cited are consistent with the delineation of visual cortical areas in our CCFv3.

Thank you for your prompt reply.

Why would you want to keep both versions in your portal, since the CCFv2 is not consistent with the literature nor with your CCFv3 and 3D viewer? This could raise some confusion. Also, the CCFv3 is very cool, but the brain areas in are not clickable.

Don’t get me wrong: The tool is amazing, I just wanted to make you aware of these glitches



I think keeping both CCFv2 and CCFv3 on our web portal has a historical reason. Before CCFv3 was built, many researcher have used CCFv2 for their publications.

2D atlas of CCFv3 is an interactive tool. It is clickable. Once you click on each structure in atlas or ontology tree, the structure will be highlighted.

Thank you for sending your feedback to us!

1 Like