NEWS.md
Prepare for first release on KWB community on Zenodo:
add vignette for checking database content and add new function import_database_content() for importing the content into R (#5)
made qmra.db public and build documentation in gh-pages (and use Gitlab only as backup)
BREAKING CHANGES: due to converting repo as R package
Moved from / into /inst/database directory (required for R Package):
MS ACCESS database (qmra-db.accdb),
folder: qmra-db_accdb (exported MS ACCESS content as .csv)
folder: rawdata (Excel files received from @PatrickSmeetsKWR and imported into MS ACCESS DB by @mrustl), see NEWS entries of prior qmra.db releases
folder: ui-workflow
(flowchart related to development of QMRA webapp user interface by @mrustl)
R files:
Moved “db-export.R” from / into R folder. In addition added Rscript .init_package.R" describing workflow for preparing this R package.
tbl_inflow:
Prepare public release (#1)
Convert repository into R package for improved documentation and also offers possiblity to add Rmarkdown export of MS ACCESS DB
(feature asked by @daniel-wicke)
Add authors: @wseis @PatrickSmeetsKWR and Aquanes as “funder”
Added a NEWS.md file to track changes to the package. Based on README.md but renamed original names to Github usernames
see https://style.tidyverse.org/news.html for writing a good NEWS.md
“rawdata” folder:
tbl_treatment:
tbl_logRemoval
tbl_treatment:
tbl_logRemoval
tbl_treatment: renamed treatment processes
tbl_references:
tbl_waterSource
tbl_ingestion
tbl_inflow:
added data from @PatrickSmeetsKWR`s EXCEL file (tbl_inflow_181008.xlsx)
new columns: mean, alpha, beta, distribution,
PathogenInReference, Notes
added only data values for mean, alpha, beta in cases these are either not to be assumed the same as min/max (i.e. for
alpha/beta) or not calcululated i.e. mean = (min+max)/2
distribution: is set to “uniform” (as default)
added raw data (used for updating DB) to sub-folder: rawdata/tbl_inflow_181008.xlsx
tbl_logRemoval
tbl_pathogenGroup:
conversion of ACCESS mdb to mySQL: https://dev.mysql.com/doc/workbench/en/wb-migration-database-access.html
Based on v0.4 after integrating most modifications performed by @PatrickSmeetsKWR in v0.5 but without rename/deletion of currently unneeded pathogens from multiple MS ACCESS tables.
tbl_pathogenGroup: added a new column “DefaultPathogenID” which contains a default PathogenID (see look-up table “tbl_pathogen”) for each pathogen group. Current defaults are:
tbl_inflow: added @PatrickSmeetsKWR`s adaptations in DB version v0.5 (but keeping all pathogen names of DB v0.4). In addition set default distribution to “uniform” instead of “lognormal”.
Reason: implementation of “lognorm” function in R package “kwb.qmra” needs to be re-programmed/tested in order to work properly (current test by myself gives unreasonable results!). Scheduled to be done after GA in Greece by @mrustl. In addition also other distributions requiring “beta” or “alpha” parameters need to be added in the R package.
tbl_health: rolled-back to table used in DB version 0.4 (containing all pathogen names)
tbl_waterSource: added @PatrickSmeetsKWR`s adaptations of v0.5 (but keeped naming for “sewage, raw”, “sewage, treated” for consistency reasons)
tbl_doseResponse: rolled-back to table used in DB version 0.4 (containing all pathogen names)
tbl_exposure: renamed to “tbl_ingestion” as proposed by @PatrickSmeetsKWR
Based on v0.4 (changes by @PatrickSmeetsKWR):
tbl_inflow Added new parameters to allow for stochastic distribution parameters to be entered “distribution”= the type of stochastic distribution to describe uncertainty about mean concentration “alpha” “beta” = the parameters of this distribution, assuming that no more than two parameter distributions will be used. e.g. with normal distribution alpha is the mean and beta is the standard deviation
@PatrickSmeetsKWR`s questions for @mrustl: align naming of distributon type with R or make sure R can translate it. Also allow for input of these parameters through the .csv file Check: entered now as lognormal, entered values as normal distributed over log concentration (mean=min+max/2 stdev=mean-min in log values)
tbl_inflow Changed Field size of min, max, alpha, beta to “Double” to allow for digits in the values (concentrations <1/l)
tbl_waterSource: changed field names for consistency and added descriptions of water sources
tbl_pathogen: added “viruses” “bacteria” and “protozoa” since this is the level that selections are made all other pathogens were removed from this table for this level assessment I saved original table: “Copy Of tbl_pathogen”
tbl_doseResponse: added viruses (rotavirus) bacteria (campy) protozoa (crypto) Note: need to check which one was chosen by @mrustl since there are multiple D-R for some organisms
tbl_health: changed to viruses (rotavirus) bacteria (campy) protozoa (crypto)
Note: Why is tbl_pathogen:tbl_doseResponse a one to many? each pathogenID has a single dose response. So these don’t need to be in separate tables Note: Log removal is now based on pathogen type, however it should be per pathogen (adenovirus different from enterovirus etc. althogh some will be the same due to lack of detail) Note: the front end selects data from databases based on user selection. It then combines all relevant information from databases in csv files for the R-tool. For the front end it keeps other data to store the scenario and later on make graphs etc.
tbl_exposure changed name to “tbl_ingestion” since exposure is generally used for the pathogens and we want to show that exposure as output. e.g. WHO QMRAdoc “magnitude and frequency of exposure to reference pathogens via the defined exposure pathways and hazardous events” although the term isn’t used consistently (e.g. Table A5.4)
@PatrickSmeetsKWR`s questions for @mrustl: the .csv file “exposure” is very different from “tbl_exposure”. Please clarify what should be sent from the front end to the R-tool:
Note that we allow only one intended use, for other applications the user needs to do a separate run of the model
Based on v0.3
Added new parameter “…Description” in the following tables:
Updated DB structure overview with new parameter
Status after internal QA with DB expert @hsonne:
ACCESS table names for each step
Based on v0.1
MS Access database “qrma-db.accdb” generated with R script “db-initialGeneration.R” containing first draft of MS ACCESS database and R script using data from Old Ford Public Irrigation scenario for generating and exporting the tables to ACCESS
All tables/queries in DB are exported with R script “db-export.R” as .csv files into subfolder “qmra-db_accdb”