User:TheLarry/Notebook/Larrys Notebook/2009/08/05

From OpenWetWare

Jump to: navigation, search
Project name Main project page
Previous entry      Next entry

Agent Based Modelling

I got the basic program done, and it seems to work fine. My rate constants are off so I'll have to tweak them to get some that look right. I'll tweak them based on the following criteria:

  1. ~100 steps/sec
  2. 1 ATP/Step
  3. ~140 steps before falling off
  4. ~1 sec before falling off
  5. backward steps are rare
  6. weird states are rare
  7. Diffusion steps take ~10 μ sec
  8. Coopertivity
    • ATP binding to front foot makes ADP binding it back foot fast
    • Unless there is an ADP in back foot front food won't bind ATP

Koch had an idea of adjusting rate constants so that an unbound state is more or less likely to bind to MT according to nucleotide state in bound foot.

Koch also wants me to include global variables so it is easier to adjust these rate constants. I am putting them in now. I am clustering the variables together so it is easier to look at. This step is just tedious, but i am guessing the next step where i constantly change rate constants will be too. But with two days left I think I can get this done.

I wrote a .vi that will graph the results. The x-axis in the graph is position. The y-axis is bound. If bound then 0 if not 1. And the color changes based on what nucleotide is bound to it. The two points are then connected with a line. I would like to add another line to signify the tail (?) but i can do that later. First i gotta get these rate constants right.

Diffusion constant symmetric but binding interact with nucleotide in other plus whether it is front or back. better chance to bind in front

OK All global variables are in place now. I gotta give them avalues now. For the record there are 132 rate constants to fuck around with. I have two days to do it. This can be tons of fun.

Oh i forgot to add in different binding to MT depending on front or back. Should fix that tomorrow.

Hyeon in Internal strain regulates the nucleotide binding site of the kinesin leading head" thinks the best time for the Kinesin to detach from the MT entirely is if there are two ADP's in both feet. Interesting. They modeled the kinesin linker like a worm like chain and then the figured there was a 12-15 pN strain put on the feet. I don't fully agree with this conclusion, but i don't know enough to have a great opinion on it. I just don't like this paper too much.


Personal tools