[ExI] Online talk next Sunday: Ken Hayworth on How to create a Connectome Observatory of the mouse brain and beyond

Eugen Leitl eugen at leitl.org
Tue Nov 15 12:44:27 UTC 2011


On Tue, Nov 15, 2011 at 12:59:24PM +0100, Giulio Prisco wrote:
> How much bigger is a human brain than a mouse brain? Like 500-1000 times?

Roughly three orders of magnitude. I expect Avogadro-scale
computing within my lifetime, so storage and processing is
about the least of my worries.
 
> Of course the mouse connectome observer is absolutely fascinating, but
> we really want to this for humans.

I see no fundamental problems with scaling up scanning
and storage/processing-wise (all assuming the resolution is
enough; the only way to know is to succeed).

The only issues I see is with sample preparation. Aldehyde
crosslinking via vascular perfusion should work, but anything
involving diffusion (soaking) is not going to work without
presectioning just after first fixation (preparation would
be identical to rodent brains after). Osmium tetroxide (lipid
bilayer fixative) via vascular perfusion might or might not 
work, heavy metal stains probably wouldn't work and I very 
much doubt the gradient dehydration and loading with monomer
and subsequent polymerization would work. At this point a literature
review and/or practical lab work is warranted.

I would like to point to this post on Mike Darwin's blog
(search for amber):
http://chronopause.com/index.php/2011/04/19/cryonics-nanotechnology-and-transhumanism-utopia-then-and-now/

...

If nanotechnology had stayed nanotechnology, instead of becoming
Nanotechnology, then it would all have been to the good. By way of
analogy, I’m not irrevocably wed to the idea of cryopreservation. I
have no emotional investment in low temperatures and on the contrary,
the need to maintain such an extreme and costly environment without
any break or interruption, scares the hell out of me. I’d much prefer
a preservation approach that has been validated over ~45 million
years, such as the demonstrated preservation of cellular
ultrastructure in glasses at ambient temperature, in the form plant
and animal tissues preserved in amber.

Buthidae: Scorpiones in Dominican Amber ~25-40 Million Years Old
[Poinar G and Poinar R.  The Quest for Life in Amber, Addison-Wesley,
Reading, MA, 1994.]

Plant Cell Ultra-structure in Baltic Amber ~45 Million Years Old:
Transmission electron micrographs of ultrathin cross-sections of the
amber cypress tissue. (a) Section of a parenchyma cell with a
chloroplast, the double membrane envelope (env), thylakoid membranes
(th) and large plastoglobuli (pg), membranes of the endoplasmic
reticulum (er), the golgi aparatus (g), the plasmalemma (pl) and part
of a mitochondrion (m). (b) Crosssection of a mitochondrion with the
outer envelope (env) and cristae (cr). (c) Cross-section of a
double-bordered pit from a tracheid-like cell with fine structures of
the primary and secondary cell walls. Size bars: (a) 500 nm; (b) 200
nm; (c) 1 mm.Cypress  [Proc. R. Soc. B272, 121–126 (2005)]

And if such an approach is ever developed, I’ll give it every
consideration, with no ego or emotional attachment to
cryopreservation.
 
> According to Ken, the required resolution can be achieved NOW (and can

I believe Ken that 5 nm voxels can be reached eventually; I'm dubious
that that resolution alone with electron microcopy data (heavy metal
stain) will be enough. This particular pudding is unconsumed yet.

> only improve). So we just need to build operational pipelines for
> preparation, readout and storage and medical research facilities (and

You can do fine with mice for the next decade at least, as far
as scanning and processing is concerned. 

> regulations or better absence thereof) able to preserve brains with
> sufficient accuracy for future readout.
> 
> On Tue, Nov 15, 2011 at 12:44 PM, Eugen Leitl <eugen at leitl.org> wrote:
> > On Mon, Nov 14, 2011 at 11:17:58PM +0000, Anders Sandberg wrote:
> >> Giulio Prisco wrote:
> >>> Ken gave a GREAT talk yesterday.
> >>>
> >>
> >> Seconded. I was amazed by it, despite both being late and thinking I
> >> knew what the state of art was. Wow.
> >
> > The 10 MUSD required for the next generation of his scanner is
> > easily the best money spent in all of connectomics. I hope Ken won't
> > have any difficulties in funding it.
> >
> > At such resolutions volumetric data storage becomes a problem.
> > With the current technology small installations beyond 10 PByte
> > will start costing some money, and a 1 cm^3 scan with 5 nm
> > resolution will produce some exabyte uncompressed. Volumetric
> > data set compression (e.g. with wavelets) would make it more
> > manageable, but will definitely have a negative impact on
> > feature segmentation and tracing.
> >
> > On the other hand, storage should become noticeably denser
> > in the next 5-10 years, so a mouse brain with that resolution
> > would be within reach.
> >
> > --
> > Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
> > ______________________________________________________________
> > ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
> > 8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
> > _______________________________________________
> > extropy-chat mailing list
> > extropy-chat at lists.extropy.org
> > http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
> >
> 
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE



More information about the extropy-chat mailing list