platelet-analysis

plateletanalysis is a Python package for the downstream analysis of platelet movement in forming blood clots.

Installation

The package can be installed on your machine as follows:

git clone https://github.com/AbigailMcGovern/platelet-analysis.git
cd platelet-analysis
pip install -e . # installs an editable version of the package

Background information

Prior to analysis, microscopy videos of growing clots, in which a small subset of platelets are labeled, are segmented to find the location of platelets. On the basis of segmentations, information is collected about the segmented objects (e.g., fluorescence intensity, volume, shape). The segmentations are then tracked using the Python package TrackPy and assigned an id (a column labeled "particle" in our data frames).

Data Configuration

Currently, the package is specialised to a particular configuration of data:

Main tracking output

The main output (format: csv) from the segmentation and tracking analysis should have the following variables:

pid – the platelet unique identifyer for that point in time
file – the name of the column with the file name (from the original ND2 video file)
xs – platelet x coordinate
ys – platelet y coordinate
zs – platelet z coordinate
... TBC

Main tracking output metadata

For each video for which tracks are found there should be an accompanying metadata file (format: csv)

path
... TBC

Collated data files

These are the data files that should be used for generating analyses using the menu interface (format: parquet). Using functions supplied in plateletanalysis you can generate one of these files (currently there are an incomplete examples of this in the examples directory - the files ending _df.py). Generally we keep one data frame per experimental condition. That is, there is one data frame per inhibitor or control. The names of the data frames that will appear as options in the analysis are specified in the plateletanalysis.config module (with will eventually be reconfigured and renamed). The path at which to find the data frame is located in the data directory in file_paths.csv. Variables include:

pid –
path –
x_s –
ys –
zs –
... TBC

Running the analysis

The user interface-based analysis can be run by calling plateletanalysis.run_analysis. Before running the analysis Please ensure file_paths.csv is properly to include the locations of (1) the directory containing collated data files, (2) the max calcium quadratic analysis regression, (3) the min calcium quadratic analysis regression, (4) the paraview files if necessary, and (5) the data containing any outliers to be excluded.

from plateletanalysis import run_analysis

run_analysis.show(run=True)

Roadmap

Ensure analyses run on a variety of machines with varying access to data
Alter config file to add functions that help the user configure to their machine (i.e., help them edit the file_paths.csv file)
Maybe move lists of treatment long and short names into a CSV (might be easier to edit)
add the t_traj_move and any related functions for mapping platelets moving in different ways at different points in time

Guidelines for version control

Here we will use git for version control.

When you want to get the latest version of the main branch of the code, ensure you are on the main branch of the git repo by entering git switch main into the command line. You can then enter git pull origin main, which obtain the main branch from the repo on GitHub (which git knows as origin). It is recommended you do this before you change anything incase the copy of main on your computer is out of date.

When you want to change the code, please checkout a new branch for the repo. Branches allow us to keep many versions of the code. When you switch to another branch, git records any changes from the main branch and allows you to switch between versions of the code. If you You can do this by entering the following command git checkout -b <new-branch-name>. If you want to find out which branch you are on and which branches exist, use git branch. If you wish to switch to another branch use git switch <branch-name>. Note that you won't be able to switch between branches unless you have staged and committed your changes (see below).

When you make changes to the code, in order to upload the changes to GitHub, you will have to first stage and commit. Staging the changes prepares them for commit. You can see which files have changes that are staged or un-staged using git status. Stage the changes by using git add <filename> or git add <filename> <filename> ... (if you need to stage multiple files). To commit the changes use git commit -m "please type a message so that we can see what the changes do". Once the changes are staged and committed, they still need to be uploaded to GitHub. To do this, you can use git push origin <branch-name>. This will provide a link to the location of the changes on the repository. In order to add the changes to the main branch you will need to use GitHub itself. Go to the link provided and click "make pull request".

jni/platelet-analysis-1