Instructions on updating the raw data that goes into Muscle DB.

Laura Hughes, [email protected] 19 Feburary 2017

All files are within the 'prep' directory

Calculate new ANOVAs

within `2017-02-19_ANOVAs.R`:

Add the additional muscle tissues to the declaration of muscle types (lines 18-22)

Update rawDataFile to be the location of the new data. Use read.csv if a .csv or readxl::read_excel if an Excel file.
Adjust numRep in the call to ANOVAlookupTable, if necessary. By default, is 6.
Re-run 2017-02-19_ANOVAs.R to calculate new ANOVAs. Files will be saved locally, where you define in script. If you get errors, check:
- there aren't extra rows / NA transcripts in the dataset
- the selection within calcANOVA.R is unique; should regex for the exact transcript shorthand followed by a single digit.
Double check results with previous runs.

Cleanup

within `importMT.R`:

change the data import to the new file

update the dplyr selection to include the new tissues for avg, SE
if adding new genes, update the transcript crosswalk file to get the new gene ontology terms.
import the newly calculated ANOVAs
Change the re-factorization to include new muscle tissues.
Copy .rds file into the data folder of the Shiny app.

Change data source in Shiny app

within `global.R`:

update the call to data (data = readRDS('data/expr_2017-02-19.rds'))

update tissueList, allTissues, selTissues
run app to double check everything works.

within `filterExpr.R`:

update muscleSymbols = plyr::mapvalues(selMuscles, ...)

Upload to server and restart Shiny server.

Copy the new .rds data file into the data directory

replace the new global.R file
deploy and restart the server.