• Question: what was your most intresting analyse youve done with an experment

    Asked by phoebee to MarthaNari, Jonny on 20 Jun 2015.
    • Photo: Martha Havenith

      Martha Havenith answered on 20 Jun 2015:

      Hey phoebee, my two favourite analysis moments so far were:
      When I looked at the task data from the mice after we switched off a pretty big part of their visual brain. I did all the computations of how often they hit the targets, how close they got to them, how quickly they noticed them… all sorts of things. And then there was this moment of ‘They don’t care at all that we switched off all those neurons! They’re still as good at the task as before!’ That was really surprising – I thought I was crazy or had made a mistake at first.
      The other moment was during my Ph.D. We were looking at really small (millisecond) delays between the activity of different neurons, and trying to find out if those delays meant something or were just there randomly. So I was writing different analyses to find patterns in those delays, and suddenly I realized that the delays all fit together, so that the neurons were active in a sequence, like notes in a melody. Sometimes a neuron would drop out here and there, but then the others would go on according to the same melody. That was a moment of really enjoying how beautiful brain activity is when you start understanding its structure.

    • Photo: Jonny Brooks-Bartlett

      Jonny Brooks-Bartlett answered on 25 Jun 2015:

      I was trying to look at what happens when X-rays are fired at proteins and I ended up with this picture: https://github.com/JonnyCBB/StructureFactorCorrelations/blob/master/test/reflectionplot12_hkl-0-2-0.pdf

      The little arrows show how the individual atoms that make the protein add together. I thought they would line up in one direction but they don’t.

      The reason this is important is because it means the big red arrow is shorter than it ought to be and this corresponds to how intense the spots on the diffraction images are (a diffraction image is the bottom image on my profile page: /lutetiumj15-zone/profile/jonnybrooksbartlett/)

      We need to know how intense the spots are to work out what the protein looks like. So this analysis has helped me understand why it’s so hard to analyse proteins.

Comments