In NiklasTR/dcp_helper: Create metadata and job files for distributed cell profiler

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

Installing dcphelper on AWS

1. push all local changes to dcphelper and the rapid project to github
2. ssh into the EC2 DCP control node on AWS
3. make sure that the control node has the ascites-biosensor s3 bucket properly mounted under bucket
4. navigate throught the bucket/ into the s3 and into /rapid
5. run git pull origin
6. run R CMD INSTALL .tar.gz

Running dcphelper on AWS

Creating metadata for brightfield flatfield correction.

1. make sure the latest version of dcphelper is installed
2. manually navigate to the XY/Image directory you want to process
3. call the following function. You have to manually specificy the channel number for the images.

Alternatively, you can also specify the path as an argument to the function. Per convention, the path has to have a trailing / and ends in the directory in which the actual images are located. It is very important to have the trailing /, as the script uses the path to determine the plate name etc.

R -q -e "library(dcphelper); dcphelper::create_flatfield_metadata(channel = 'ch2')"
# Be careful with the "" vs. '' quotes. 

The function runs in mutliple steps.

1. List the content of a directory and store the list. This can take a lot of time, but only needs to be done once. (Important: if you update the files in the directory this will not change the filelist automatically. You can overwrite this behvaiour by calling the function with the argument force = TRUE).
2. Access the filelist and filter the images of interest
3. Format the metadata table so it can be processed with cellprofiler.

The output of the function is located in ~/metadata and has the name of the directory in it. You have to manually make a copy of the file and rename it to metadata_bf.csv so it gets processed by cellprofiler as of now. This might change in future versions of the pipeline.

To-Do explain below:

metadata <- dcphelper::create_flatfield_metadata_split(channel = 'ch2')

NiklasTR/dcp_helper documentation built on June 12, 2020, 10:22 p.m.