Notes from the last DANG meeting.

-phyloseq R package to load microbiome data into R from mothur output -import.mothur, wrapper for a bunch of other functions in R… what’s the vegan package

-code sharing: github–public vs. private repositories -R.md file with the output visible (graphs, results, etc)… the code can be hidden or visible for the PI

-make.sra–to submit sequences to NCBI and make publicly available in the sequence read archive -have to fill out the mimarks table for it

-dplyr package updated from the plyr package–for wide/long format changes of the data, transforming the data in that way, reshape function -tidydata–made by the same people as plyr -stringr–supposed to be good for working with strings -sometimes excel/text wrangler manipulations are just easier to work with metadata to clean it up and organize it -but always save the raw data file -broom package -90% of the work sometimes is getting the data into one organized consistent form

-“?files” tells you what files are there… look at the files help page -in Rmd can have a chunk of code that is python/r/terminal code

-ggplot2 vs base commands in R–visuals are cleaner perhaps, but difficult to learn how to use, although some people say the opposite -base commands–might be a lot more parameters to tweak but can totally control everything in the graph and save it for the next time you need to make that graph -measles infographic to compare the two–look it up, how people made the same graph in different ways

-R style guide recommendations, be consistent -sometimes its nice to have examples of input files to show what that is supposed to look like, also nice to have an example of the output to refer to

-disadvantage to using rstudio is if need to use the cluster then should use the terminal