Working Projects

[iamciera]

Tasks to work on:

• Displaying both video together. Displaying two channels with two different colors?
• Working on automating threshold
• Work on tracking spots through time
• to reduce noise track spot through time and only keep ones that last multiple z-stacks (possible strategy to reduce noise)
• prepare questions for meeting with scikit image people
  Start new notebooks to tackle topics (edited)

[iamciera]

Update November 14

• Automating thresholding could be a big challenge
• Samantha worked on new blob algorithms and coordinate detection.
• Look into reducing the noise.
• Use the nuclei location to track spots? - @iamciera look into using blob for this
• look into Weka segmentation either in fiji or possibly in Python?
• look into creating functions for code - @satang1

[iamciera]

Update November 21

• Explore the difference of gaussian results closer.
• Test on another z-stack
• Start exploring how to represent on plot (time, location, on/off)

[iamciera]

December 12 - Overbreak task

• Try a few algorithms to identify MS2 spots (little spots)
• Main Goal: Pick one algorithm and create a dataframe that has location of spots. Columns: X, Y, area?, SD, timepoint (each Z-stack). We want this information for 1. to understand algorithm output 2. possible use timepoint of spots through time to identify if spot exists 3. Quantify if the different algorithms are better or worse than each other. 4. Set dot movement to identify if dot is the same.

[iamciera]

@ciera get nuclei (larger spot) location and most importantly area. Again, because there can only be one MS2 spot (smaller spot) per nuclei. This could be away to reduce noise.

[ciera]

FYI @iamciera, when you use the @ before a username, it notifies the person with that name. You definitely don't want me getting location and area of nuclei. :)

[iamciera]

Oh hello fellow @ciera! I don't meet many of you in real life. Thanks for the heads up.