Yeast Time Series Analysis
Data analysis of yeast time series to discriminate between cells based on position of associated patches. Characterization of patches via intensity and distance profiles.
Analysis is subdivided into five phases:
- F1 Cell Profiler pipeline is used to detect and track cells and detect patches. Pipeline needs access to original movie. Results are saved to two CSV files corresponding to objects in interest (FilteredCell, Patch).
- F2 Use script
cell_filter.pywhich outputs a file containing rules for cell classification (rules.txt). We use rules to filter out bad cells (those moving, shaking, growing, without patches etc.). To construct rules, script
cell_filter.pyneeds both CSV files from phase 1.
- F3 Cell Profiler pipeline is used to filter movie based on the rules from phase 2. Pipeline needs access to original movie and rules for filtering cells. It outputs a sequence of images with bad cells removed.
- F4 Cell Profiler pipeline detects and tracks patches. Its input is a movie with filtered cells (C1-006009001-corr-filtered.tif) produced in phase 3 and separately first image of the movie (C1-006009001-corr-filtered-1.tif). The latter is used for xy alignment of images. Pipeline produces two CSV files with measurements for cells (TrackedCells) and patches (TrackedPatches).
- F5 Use script
patch_tracking.pyto plot graphs with intensity and distance profiles of patches, various alignment (start, end, max) of profiles, save file with statistics (patches.txt, 1C_patch_tracking.csv). Paths to both CSV files from phase 4 must be passed to script.
Developed during visit at University of Toronto, CCBR; August--October 2012.