Visualized Transforms


What you are seeing is a series of transforms from an amazing piece of code, Viking that was created by James Anderson which allows us to do our connectomics work.  We complain quite a bit about the tools we use to perform our science, but the reality is: Viking is the best annotation tool in the world for elucidation of neural circuits in connectomics work.

The transforms from this visualization are the derived transforms from the first 59 slices of several hundred 1200 tile electron microscope mosaics that have been registered to one another.  These registered mosaics allow us to track neurons through these large multi-terabyte datasets, keeping track of all of the synapses and gap junctions among other things that allow us to derive circuitry.  I was able to create this animated gif by looking at the transforms independent of the TEM tile data while we were waiting for a RAID array to rebuild after swapping out a failed hard drive in the server array.  On days like this one, I wished that we could afford to upload our couple hundred terabytes to a cloud for someone else to manage.  At least I got some art out of it while we were down…

One Reply to “Visualized Transforms”

  1. Your link to James Anderson had a heading on the right called connectomics and eternity. There were five slides by Robert Marc in that section. The first discussed the difficulty of computing the brain’s possible networks with a Cray Titan, but in the fourth slide his last point was that brain data-flow rates are slow. Will you please explain what slow means in regard to data-flow rates and why that is a difficulty for downloading or uploading brain contents.

    If the brain has slow data-flow rates I assume the retina does also yet it still seems to be very difficult to understand. When I think of slower I think of something easier to understand yet that does not seem to be the case with the brain or retina. I appreciate your patience with my questions. Understanding Photoshop is a whole lot easier than understanding neuroscience.

Leave a Reply

Your email address will not be published. Required fields are marked *