Processing Fluorcam data into Tcrit, T50 and Tmax

Based on processing steps outlined in Arnold et al. 2021 and refined by Madeline Moran, M.S., this repository provides a workflow for automating detection of thermal limits in high-throughput chlorophyll imaging fluorescence.

How to use this repository

1. Clone or download to the desired location on your local computer

2. Click on fluorcam-processing.Rproj, which will open the RStudio project in a new window

3. Run renv::restore() in the console to load package library (see here for details about collaborating with 'renv')

4. From the file pane, open workflow.R

5. Run the first 25 lines of the script, which will load the requisite packages and create the data folders

6. Move a Fluorcam file (.txt) into data_raw/; rename if needed. I recommend keeping the YYYYMMDD as the first part of the file name so files will be chronologically ordered.

7. Move the accompanying well labels file (.csv) into data_labels/; provide with the identical file name as the Fluorcam .txt. The initial column should be labeled well and consist of wells from A1, A2, ... H3 (n = 45). One or more additional custom columns can be specified, e.g., site, species , sampleID, etc.

8. Run the remainder of workflow.R. When complete, data_processed/ should contain one folder and one .csv file of the same name as the data_raw/ input:

   • The folder will contain a .png image for each well location (A1,A2, ... H4,) plotting the raw and fitted data with thermal limits (Tcrit, T50, Tmax).
   • The file will contain the thermal limits for each well location in tabular form

Development notes

This processing workflow is in development. Current features include:

• detecting outliers in the raw fluorescence data (removes points exceeding the median + 3*IQR)
• rescaling fluorescence between minimum and maximum, but only designating minimum if occurred prior to maximum
• calculating T50 only if it occurred prior to Tmax

Desired features still include:

• adding conditional to breakpoint for loop if model does not converge

• adding date_run and date_processed columns to final output - is date_run available as part of the Fluorcam file name?