Revisiting the manuscript

We got comments back from other co-authors so it’s time to dive back into this project! To ease myself in, I started reorganizing my files in the repository and creating a README map. It’s a low priority task that’s much easier than actually addressing comments on the text…so here we are.

I picked up where Shelly left off with the README map by assigning categories and number keys to all scripts in the code README.md. I designated some scripts related to DMC genomic locations to an “exploratory” status and tagged those with a “100” prefix. Although there are two separate categories for methylation characterization and genomic location, my scripts for these categories are combined. I added that information in the README, but it may be useful at some point to separate those scripts. Again, low priority.

Once I was finished with the code directory, I moved on to output. We decided to split the output directory into three subdirectories: figures, supplemental material, and intermediate files. All the subdirectories that were once in the “analyses” folder needed to be moved and organized! I first went through my intermediate files and ensured these files were actually useful. I deleted a few that were truly intermediate files that could be generated by code, but wouldn’t be too important to access immediately. Then, I renamed the directories with the category tag as the script that produced them, and updated the README.md. I also added the same organizational structure to the intermediate file README.md that the code README.md had, with breaks for each of the headers in the manuscript. Finally, I updated the figure and table links in the figures README.md and the supplemental material README.md.

Feeling motivated to do something that was a higher priority, I tagged Katie in this Github issue to get her up to speed with the various decisions we made before selecting a statistical approach. She asked why we didn’t use a contingency test for our data, which was something we previously considered before we settled on multivariate and linear model analyses. Hopefully she can suggest a better statistical approach so I can clarify my manuscript sections!

Going forward

1. Determine a valid statistical approach
2. Recharacterize CpG methylation based on Katie and Jon’s comments
3. Address all other co-author comments