Allen connectome

24 views
Skip to first unread message

Kai Chuang

unread,
Aug 21, 2021, 2:31:06 AMAug 21
to TVB Users
Hi,

I'm trying to simulate the activity of the somatosensory pathway from the thalamus to the cortex in mouse. For that, I will need to separate the nucleus, VPM, that projects to the S1 whisker, area and the other nucleus, VPL, which projects to the S1 limb area. 

I used the Allen connectome builder to create the long range connections. However, no matter how I set up the parameters, I could only get a rather big "lateral posterior nucleus" of thalamus that include the VPM, VPL and other nuclei. 

I tried: 50 micron resolution, injection volume 10% and minimum size 1000000 um^3 (both VPM and VPL are ~500x the minimum size). Please advise. Thank you.

Spase Petkoski

unread,
Aug 21, 2021, 1:41:33 PMAug 21
to tvb-...@googlegroups.com
Dear Kai,

Allen Connectome Builder can be only as fine-grained as the experimental connectivity obtained by the tracing done by the Allen Institute. 
As you can see in the Interactive Atlas Viewer 
there is no further division of "lateral posterior nucleus" in other nuclei. 
If the Allen connectome gets some updates with new experiments, they should be also accessible by the Allen Connectivity Builder in TVB.

Best regards,
Spase



--
You received this message because you are subscribed to the Google Groups "TVB Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tvb-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tvb-users/8523a460-965e-40ca-bad7-ea05c9f626bdn%40googlegroups.com.

Kai Chuang

unread,
Aug 21, 2021, 6:01:12 PMAug 21
to TVB Users
Dear Spase,

Thanks for looking into this. Sorry about the typo. I was referring to the "Posterior Medial" nucleus of thalamus, as shown in the screenshot below. It looks like correspond to the "ventral posterior complex", which includes the VPM, VPL that I need, in the atlas: http://atlas.brain-map.org/atlas?atlas=1#atlas=1&plate=100960092&structure=709&x=5119.125&y=4063.75&zoom=-3&resolution=16.75&z=6

Screen Shot 2021-08-22 at 7.53.49 am.png

spase.p...@gmail.com 在 2021年8月22日 星期日上午3:41:33 [UTC+10] 的信中寫道:

Spase Petkoski

unread,
Aug 22, 2021, 4:04:59 AMAug 22
to tvb-...@googlegroups.com
Hello Kai,

Alright, it seems like there are injections in that region 
Maybe the criteria of "injection volume" isn't fulfilled for these projections to be included? 
I will also check with others working more recently with the Allen connectivity builder.

Best regards,
Spase


Kai Chuang

unread,
Aug 23, 2021, 2:09:31 AMAug 23
to TVB Users
Hi Spase,

The volume of VPM is ~800000000 um^3 (0.8 mm^3), and VPL ~430000000um^3 (0.43 mm^3). The tracer injection volumes in the Allen brain are from 0.1 to 0.6 mm^3 for the VPM, and 0.17-0.2 mm^3 for the VPL, which correspond to > 12% for VPM and > 40% for VPL. As the tracer spread over multiple areas, this injection ratio will be lower (>8% for VPM, >12% for VPL) when considering the proportional spread. That's why I set it to 10%.

spase.p...@gmail.com 在 2021年8月22日 星期日下午6:04:59 [UTC+10] 的信中寫道:

WOODMAN Michael

unread,
Aug 23, 2021, 2:13:24 AMAug 23
to tvb-...@googlegroups.com

Hi,


When writing the Allen connectivity builder, we tried to include some heuristics to avoid imbalance in the volume of different regions.  The ontology that Allen provides is also not an atlas per se, but a tree, and it changes over time.  These factors likely contribute to the missing region here.  We can have a look but if you are able to take a look at the implementation of the builder workflow


https://github.com/the-virtual-brain/tvb-root/blob/master/framework_tvb/tvb/adapters/creators/allen_creator.py#L155


and the functions that follow there, you may spot something.


cheers,

Marmaduke


From: tvb-...@googlegroups.com <tvb-...@googlegroups.com> on behalf of Spase Petkoski <spase.p...@gmail.com>
Sent: Sunday, August 22, 2021 10:04:21 AM
To: tvb-...@googlegroups.com
Subject: Re: [TVB] Allen connectome
 

Kai Chuang

unread,
Aug 23, 2021, 8:13:52 PMAug 23
to TVB Users
Hi Marmaduke,

I wonder whether it is the annotation (and its volume) or the structure_tree insufficient so that the sub-structure was ignored.
        vol, annot_info = cache.get_annotation_volume()
        structure_tree = cache.get_structure_tree()
I inserted a print(annot_info) after the above commands, but couldn't see the output either from GUI or command-line mode.

marmaduke.woodman 在 2021年8月23日 星期一下午4:13:24 [UTC+10] 的信中寫道:

WOODMAN Michael

unread,
Aug 24, 2021, 3:54:07 AMAug 24
to tvb-...@googlegroups.com

Hi


I think that's the right path, but it would probably be easier to run with a debugger to inspect the tree structure.  In the case of a print statement, the output is showing up in the log files (in $HOME/TVB/logs).


cheers,

Marmaduke


From: tvb-...@googlegroups.com <tvb-...@googlegroups.com> on behalf of Kai Chuang <kaic...@gmail.com>
Sent: Tuesday, August 24, 2021 2:13:52 AM
To: TVB Users
Reply all
Reply to author
Forward
0 new messages