stanstrup / qc4metabolomics Goto Github PK

funs() --> list()

• automatic msconvert (Proteowizard I suppose? Double check). Batch script on a schedule? File monitor?
• Automatic copy of converted file.
• copy sample name and project name from samplelist name from the raw data.

Settings app

• move settings in a separate app that can always run

Control panel that can

• reschedule all files
• reschedule selected files
• reevaluate identical files --> refer to the ones that still exist
• prune missing files

6000 cut-off check. perhaps the min-int / RT plot

Right now we use correlation.
I think something like euclidean, but normalized, would be better. What can we use?

There is a lot of duplicated code for adding to the ignore list. This should be a function in the package.

Sorry for the elementary question, but I am not very familiar with MySQL and I cannot figure out how to run the Shiny application.

I believe I am not properly setting up the MySQL connection so the culprit is there. I would really appreciate any feedback to help me understand what I am doing wrong here.

I receive this error:

Warning: Error in .local: could not run statement: Table 'metabolomicsSQLDB.std_compounds' doesn't exist
Warning: Error in .local: could not run statement: Table 'metabolomicsSQLDB.file_info' doesn't exist
Error in .local(conn, statement, ...) :
could not run statement: Table 'metabolomicsSQLDB.std_compounds' doesn't exist
Warning: Error in .local: could not run statement: Table 'metabolomicsSQLDB.file_info' doesn't exist

These are the steps I have taken to set up the Shiny app:

First, in Ubuntu terminal.

$ sudo /etc/init.d/mysql start
$ sudo /usr/bin/mysql -u root -p
    mysql> CREATE DATABASE metabolomicsSQLDB;
    mysql> GRANT ALL PRIVILEGES ON *.* TO 'R_user'@'localhost' IDENTIFIED BY 'pass1234';

Then, in RStudio.

devtools::install_git("https://gitlab.com/R_packages/massageR.git")
devtools::install_github("stanstrup/QC4Metabolomics/MetabolomiQCsR")
system('wget https://github.com/stanstrup/QC4Metabolomics/archive/master.zip')
system('unzip master.zip; rm master.zip')
MetabolomiQCsR.env$db$db <- "metabolomicsSQLDB"
MetabolomiQCsR.env$db$user <- "R_user"
MetabolomiQCsR.env$db$password <- "pass1234"
MetabolomiQCsR.env$db$host <- "127.0.0.1"
runApp(appDir = "QC4Metabolomics-master/Shiny App/", launch.browser = T)

Am I missing something? Thanks.

use validate or need in shiny I think.

DB with last script run time --> run script every X minutes. Need to be able to execute the script without waiting for results.

• Find way to merge single file XCMS objects.
• Find list of standards.
• RT dev
• mz dev
• Intensity compare

• do one line per sample. not per date. the put the days where they need to be.
• make it long enough to read all contaminant names

I am trying to install MetabolomiQCsR in R 3.4.4 in the following way:

devtools::install_github('stanstrup/QC4Metabolomics/MetabolomiQCsR')

I get an error stating that massageR is missing.

ERROR: dependency 'massageR' is not available for package 'MetabolomiQCsR'

I cannot find massageR anywhere.

Need to check of it does now but it should do some filtering on the sql side.
Like

• Only entres above the min setting --> get unique compounds
• pull only those compounds

• Define database structure
• Select database
• Select R <-> database linker

there should be a new function now to do that. Need to find it.

heatmap punch card

EDIT: Done in 35ca533

Can we group the features some smart way?
Cut the dendrogram?
Different measure of distance?

bnosac/cronR#18