To run via command line: python StatisticalModel_DenticleOrganization.py <name of input file> <number of iterations>
where the input file is a csv file (*_CellbyCell.csv), structured as described below.
For more details, see Spencer, Schaumberg and Zallen, Molecular Biology of the Cell, 2017. If this code has been helpful to you, please cite our paper.
Written & tested primarily in python 3.5 (Anaconda). Runs in python 2.7, but has not been extensively tested.
- numpy
- pandas
- scipy (scipy.stats)
- matplotlib
- seaborn (see requirements.txt)
To run via command line: python StatisticalModel_DenticleOrganization.py <name of input file> <number of iterations>
Takes the output *_CellbyCell.csv from the MATLAB script set "Analyze Denticle Organization.m".
format:
| column | label | description |
|---|---|---|
| 01 | embryoID | unique # for each embryo, so that the script finds individual embryos, and so I can trace problems; embryo pattern is 0xxx |
| 02 | row | aka cell column; [1:6] |
| 03 | belt | 'band' number, corresponding to the abdominal segment number; I focused on denticle belts 3-7, as these are most similarly structured. |
| 04 | cell | from left to right (embryo tail-up) (or top to bottom, embryo with head pointing left) , the # of the cell within that specific belt/row (e.g. if there are 6 cells in that row/belt, there will be 6 rows of data, with this column [1:6] |
| 05 | dentincell | # of denticles in that cell |
| 06 | Dvlen | length of the cell |
| 07 | dentEdgeL | distance between the left/top-most denticle and the left/top-most cell DV edge |
| 08 | dentEdgeR | ditto for right/bottom-most |
| 09 | AdjL | for a given cell, the distance from the left/top-most denticle in that cell to the closest denticle in the adjacent cell moving left/up |
| 10 | AdjR | ditto but for right/down |
| 11 | Intra2 | distance between denticle 1 and denticle 2 (from left to right) |
| 12 | Intra3 | distance between denticle 2 and denticle 3 |
| 13 | Intra4 | distance between denticle 3 and denticle 4 |
| .. | ... | etc |
- choose number of repeats/iterations
- choose to use the 'absolute' cell DV length or the summed total of denticle-edge and denticle-denticle distances (as in the equation below). To switch between these options, comment/uncomment the sections of code under the headers.
- using the 'absolute' DV length (dist between edge markers)
- using the summed DVlength (sum of dent-edge, dent-dent ... dent-dent, dent-edge); sum to get the additive, rather than absolute, DV length in StatisticalModel_DenticleOrganization*.py
Basic methodology:
The equation L = α*D + D(N-1) + α*D describes the general case for determining the distance between denticles (D) given the known cell length (L) and the number of denticles (N) in the cell, where α is the spacing ratio (the ratio of the average denticle-to-cell-edge distance to the average denticle-to-denticle distance).
Written by Alison Spencer in Jennifer Zallen's lab at the Sloan Kettering Institute.