906
edits
Changes
Wrote about uploading data
* Make a new branch called '''something-manuscript''' to describe the manuscript you are working on. You will work only on that branch (not on master) until the paper is ready to submit. You can see what branch you are on in Github Desktop by looking on the top bar where it says "Current Branch"
* Create a new folder for your manuscript and separately for your code. Use the existing directory structure to understand where to put your data.
* In the manuscript folder create a word document called manuscript and another illustrator document called figures. These will become the files you will submit. The word document can start with headers for Abstract, Introduction, Methods, Results, Discussion References and Figure/Table Legends.
* In your manuscript folder create a file called README.md using a text editor such as Sublime or Textwrangler. This is where you will note where to find the data that goes into each figure panel. The format of this file is markdown, and an example can be found [https://raw.githubusercontent.com/BridgesLab/TissueSpecificTscKnockouts/master/Manuscripts/Muscle%20Tsc1%20Knockout%20Paper/README.md here].
* Also create a markdown file in that folder called TODO.md. Use this to make little notes to yourself as you are writing (''i.e.'' check on stats for this experiment, or look into more papers on such and such).
This is a recommendation, but there are other ways to approach writing a manuscript
=== Upload and Write About the Data (Figure Legends/Methods/Results Section) ===
* For every experiment, you should take the script from the shared drive (Rmd file) and its associated data file and put it in your code folder. Remove anything in the Rmd script about updating data from mousedb, as we can update that later if needed. The script should be able to run using only the datafile. You may want to put datafiles in a separate folder from scripts in your code, if so you will need to update the location of the datafile in your Rmd file. Check that your Rmd file still runs by clicking '''Knit to HTML''' (it should create a md file, an html file and a figures folder with all the figures). If there is a compilation error check that the datafile is able to be found correctly. Check the names on the experiment, and make sure everyone who participates is included in the author list.
* Once the script/figure runs modify the panel if needed for formatting. See [[Generation of Figures in Illustrator]] for some details and suggestions.
* Find and drag the generated pdf file (in the figures folder) to its location on a whiteboard in the illustrator file you created. Each whiteboard should be a single figure (''i.e.'' Figure 1), often with multiple panels.
* Update the README.md file with the name of the script, the name of the figure panel and expected location in the illustrator file (''i.e.'' Figure 1C).
* Describe the figure in the Figure Legends section, make sure to note that statistical test being used and the sample size (if your Rmd file does not have a table showing the n create on). Mark p-values with asterisks on the illustrator file
* Describe the experiment in the Results section. Make sure to explain why the experiment was done, what was found and include both effect sizes and p-values if pertinent.
* Describe the method, if not already present in the methods section. Our standard is to provide sufficient methodological detail for all experiments except those using kits where you can refer to the manufacterers instructions. Make sure to note the catalog number and source of all key kits/reagents and add RRID's where available. Make sure to add references to key analytical tools and methods.
* Repeat with the next experiment.
* When you are completed you should have a full methods section, a completed set of figure legends and a complete (although likely disjointed) results description
=== Write the Discussion ===
=== Write the Introduction ===
=== Write the Abstract and Acknowledgements ===