In summer of 1997 I interned at the Santa Fe Institute. Barry McMullin was there as well, using swarm (an early cellular automata library) to reimplement and extend the original autopoiesis algorithm. His report: https://www.santafe.edu/research/results/working-papers/comp...
Late to the party, but corresponding author on the paper here. Cool to see this on HN!
We have less than "half the picture" here. Not just weights; also missing electrical synapses, neurotransmitters, etc. We also don't know the spatial scale of neuronal arbor integration. Furthermore these are just the image data, not the complete connectome; people still have to trace circuits by hand in this dataset. Collaborators are starting to crack the segmentation problem, but it is still early days.
"URL to this view" lets you share URLs to whatever you're looking at.
----
edit:
in mammals there is pretty good circumstantial evidence that post-synaptic density size correlates with evoked postsynaptic potential, but this hasn't been clearly and directly calibrated yet, and could vary from cell type to cell type
What technology needs to be developed to get the data on synapses and neurotransmitters? A high-resolution Raman imaging microscope? And is a dead brain sufficient, or would you need a real-time noninvasive scan of a living one?
In the fruit fly, neuronal cell types are highly morphologically stereotyped and identifiable across animals. This means that for a given cell type, you can collect data on electrical synapses in animal A, on transmitters in animal B, and on electrophysiology in animal C, and in this fashion assemble a unified, multimodal view of the parts involved. Our whole brain EM volume lets you see how those parts are connected.
In the above examples dead brains are okay except for electrophysiology, where the brain needs to be alive.
Hey cool, 3rd author Shane Gonin is now at Janelia (where my lab is) and I know a bit about this work. Funny to see it pop up here.
The kind of electron microscopy (EM) of brain tissue I do relies upon embedding the tissue in a resin called Epon. Epon has excellent cutting properties and low intrinsic contrast in EM. But in order to embed tissue in Epon it has to be completely dehydrated, which quenches genetically expressed fluorphores like GFP.
My fantasy for these genetically expressed buckyball-like proteins is that one could engineer their interior to be sufficiently hydrophilic that GFP fluorescence would survive complete dehydration of the surrounding tissue, instead relying on the polarized residues of the amino acids in the interior. This would let us combine highest quality EM with highest quality light microscopy in the same sample -- which would be very useful indeed.
I doubt you could avoid diffusion in these proteins to the point you desire. It would be worth looking into species that exploit cryptobiosis, specifically anhydrobiosis (rotifers, tardigrades, daphnia, and C. Elegans are all examples). In most cases they can dessicate extremely, some of them by exchanging glucose or water with trehalose (itself a fascinating topic worth further study), so i wonder if you could make trehalose binding sites inside the body.
"The day will come when David Bowie is a star and the crushed remains of his melodies are broadcast from Muzak boxes in every elevator and hotel lobby in town."
Bock lab, Janelia research campus, Howard Hughes Medical Institute (HHMI) | Ashburn, VA | ONSITE
Data scientist / Neurogeometer
We recently finished acquiring an electron microscopy dataset encompassing the entire brain of a fruitfly (120 million images, 115 TB on disk, voxel size 4 x 4 x 40 nanometers). A team of tracers is manually skeletonizing portions of this animal's nervous system, and we are starting to see interesting circuit motifs emerge.
We are looking for someone to help build and use tools supporting this analysis (neurogeometry), as well as help wrangle the millions of images continuing to flow from our three microscopes (~500 TB of primary image data anticipated in the next two years). During this time we expect to switch from RAID-based file systems to an object store (likely Scality), with substantial support from a team at Janelia outside the lab.
Desirable attributes:
- Education: B.S. or M.S. in Computer Science or related field (or an equivalent in education and experience)
- Strong familiarity with Linux, including command line wizardry and detangling dependency nightmares
- Proficient in Python and at least one other programming language (e.g Java or C/C++)
- Experience with utilizing SQL and/or NoSQL data stores (e.g. MongoDB, PostgreSQL, Microsoft SQL Server)
- Experience with image processing (ImageJ, SciPy, Matlab, OpenCV, etc.)
- Comfortable working with new programming languages and tools when necessary
- Familiar with RESTful web services
- Experience using an HPC cluster (MPI, SGE/OGE, Spark, etc.)
- Follows current trends in industry and academia
- Willing to get hands dirty with hardware
- Ability to concentrate in a fast paced and dynamic environment
- Ability to communicate effectively, orally and in writing
- Ability to take initiative, prioritize tasks, use good judgement and monitor completion of assigned duties.
- Experience or strong interest in working in an interdisciplinary environment.
As a grad student back then, I saw a pyramid scheme. Training programs were (and are) funding way more students than there is room at the top. If funding at the top level stops expanding, the current scenario unfolds.
If I have undergrads in my lab who want to go on to graduate school, I give them sufficient information on these numbers that they can provide "informed consent". If they still want to go ahead, they receive my full support.
Given the hosts and the venue, the best topic to bring up could be software patent reform. This is something pg has written about in the past [0] -- I wouldn't be surprised to learn this issue is the driving reason for hosting the dinner.
Fascinating idea: insane society-scale bureaucracies make puzzle solvers out of everyone.
Sokoloff was Russian, born and raised in a city on the Volga River. He had an explanation for why so many of his countrymen wound up in high-frequency trading. The old Soviet educational system channeled people into math and science. And the Soviet-controlled economy was horrible and complicated but riddled with loopholes, an environment that left those who mastered it well prepared for Wall Street in the early 21st century.
And a later study by him: https://pubmed.ncbi.nlm.nih.gov/15245628/