Icon

PrecisionTox Concentration-response analysis Reporting Template BMD 22mar2024 KP v2.10

PrecisionTox Concentration-response analysis (log-logistic model)

Changes from 2.9
- Avoid failing when no notes are provided
- Adjusted for R version 4.3 and changes in bmd codes
Changes from 2.8
A compliance check of the reporting template is conducted initially. If your table has dfferent columns names (case sensitive!) and or columns are missing, a list of errors is provided.

Changes from 2.7
Fixed a but when no p-values for trend test could be generated
(in case of all effect values = 0)

Changes from v2.6:
Adopted workflow so that it can handle data and tables with no effects in any concentration (on the expense that it is a bit slower than previos version).

This workflow uses data from an input file compliant with a dedicated PrecisionTox template
with the same version number and will automatically generate concentration-response modelling data and a reporting file with raw and modelled data.

# PT_BMD_summary file naming

the format of the file name is: "PT_BMD_summary_(species)_(four-digit-number)"

for each of the six organism used in PrecisionTox, the reporting template file is numbered from 0001 onwards, that is, 0001-9999. The number represents a consecutive number that reflects a batch of analysis and does not relate the the PrecisionTox chemical ID.

examples of the PT_BMD_summary file names: "PT_BMD_summary_hsa_0001.xls", "PT_BMD_summary_dme_0011.xls", "PT_BMD_summary_dma_0250.xls"

All data must be in a form that values increase with higher concentration (in case the compound is provoking an effect). Hence, you may have to transform data in advance (coding would be very demanding to allow increasing and decreasing data types).

Data that can be modelled by a Hill (log-logistic) equation, are identified by comparison of the AIC to a linear model with slope 0.

Furthermore, the percent variance reduction is calculated as an intuitive indicator whether the curve fitting is approximating the data.

Note: It is important that you upload a table with the same structure as in the template. It is possible to only upload one experiment or one replicate.

To check for compliance of workflow and template refer to the appropriate version number at the end of the file name and workflow. E.g. template files and workflows ending with v2.3 are compliant which each other.

Template file location on sharepoint:
Documents\General\Data Management Plan Working Group\Metadata_Templates

stefan.scholz@ufz.de












Models (only Hill-slope has been used)• Linear:Linear (“fitMod”): y = E0 + δx E0 … intercept, δ … slopeLinear in log (“fitMod”): y = E0 + δ log(x) E0 … intercept, δ … slope• ExponentialExponential decay („drm”): y = c + (d - c)(exp(-x/e)) c,d … upper and lower limit, e …steepnes of decay/growth Exponential Model ((“fitMod”): y = E0 + E1(exp(x/δ) - 1) E0 … placebo effect, E1 … slope, δ … Parameter, controlling the convexity of the model• PeakGaussian („drm”): y = c+(d-c)*exp(-0.5 * ((x-e)/b)^2) c … background effect, d... peak, e … critical point, b … growth rateLog-gaussian („drm”): y = c+(d-c)*exp(-0.5 * ((log(dose)-log(e))/b)^2) c … background effect, d ... peak, e … critical point, b … growth rate5P Gaussian („drm”): y = c+(d-c)*exp(-0.5 * ((x-e)/b)^f) c … background effect, d... peak, e … critical point, b … growth rate5P Log-gaussian („drm”): y = c+(d-c)*exp(-0.5 * (sqrt(((log(x)-log(e))/b)^2))^f) c … background effect, d ... peak, e … critical point, b … growth rate• Sigmoidal4P Logistic (“fitMod”): y = E0 + Emax / (1 + exp ((ED50 - x) / δ)) E0 … left-asymptote corresponding to a basal effect level, Emax … Asymptotic maximum effect, δ … steepness of the curve4P Log-Logistic; Hill (“fitMod”): y = E0 + Emax (x^h / (ED50 ^h +x^h)) E0 … left-asymptote corresponding to a basal effect level, Emax … Asymptotic maximum effect, h … steepness of the curve (Hill parameter)Gompertz (“drc”) : y = c + (d - c)(exp(-exp(b(x-e)))) Weibull type I („drc“): y = c + (d - c)exp(-exp(b(log(x)-log(e))))Michaelis-Menten (“fitMod”): y = E0 + Emax (x / (ED50 +x)) E0 … placebo effect, Emax … Asymptotic maximum effect To change symbols in plots selectsymbol numbers from e.g.http://www.endmemo.com/program/R/pchsymbols.php Concentration-response modelling single or multiple replicatesusing PrecistionTox templatesigmoidal, log-logistic, Hill-slope modelConstraints for min and max are set to 0 and 100. The loop design allows to group files with replicates of the same chemical and toanalyse several endpoints simultanously. Note that the workflow may need adaptation in case the file name structure and data tablestructure is changing.stefan.scholz(at)ufz.de Node 696Save a temporary fileRemove columnsnot required for modelingAdjust alsoempty table configurationto align with output ofmodel nodeSelect symbolsNode 974Node 977Node 978Modelled curvesNode 1075Node 1114Node 1115Node 1195Node 1196Node 1198Include "yes"Convert concentration, replicateand effect to number (optional)Convert INTto DOUBLEExtract experiment IDNode 1220Node 1222Node 1224Definitionof parametersDeleteempty rowsSelect also Hillparameters Raw dataNode 1236To replace group loop startfor improving speedNode 1255Node 1256Node 1257Node 1258ReportingNode 1260Node 1269Node 1273Endpoint explanationNode 1276Node 1277Node 1278Node 1282Node 1283Node 1290Node 1291Node 1293optional row filterfor testingspecific data setsCalculate effect threshold(min+10)Node 1298Node 1299Node 1300Node 1301Node 1302Node 1303 Collect metainformation Excel Writer (XLS)(deprecated) Donwload Excelresult file Column Filter Conc-resp modelling Table Creator Table Creator Catch Errors(Data Ports) Try (Data Ports) XLS Sheet Appender(deprecated) Create TempDir (legacy) Number To String String Replacer Math Formula String To Number Column AutoType Cast View results Row Filter String To Number Column Rename GroupBy Joiner (deprecated) Missing Value Column Resorter XLS Sheet Appender(deprecated) Row Filter File upload andparameter selection XLS Sheet Appender(deprecated) GroupBy Grouping Remove values fromselected experiments View results Joiner (deprecated) Row Filter Prepare model parametersto join with raw data Sorter Excel Sheet Appender(XLS) (deprecated) Missing Value Column Resorter Arrange images XLS Sheet Appender(deprecated) Column Rename Column AutoType Cast Column Resorter Inspectuploaded data Joiner Table Rowto Variable Chunk Loop Start Loop End Math Formula Automatically flagand remove BMDs Row Filter Math Formula Table Validator CASE Switch End Excel Reader Excel Reader Excel Reader String To Number Models (only Hill-slope has been used)• Linear:Linear (“fitMod”): y = E0 + δx E0 … intercept, δ … slopeLinear in log (“fitMod”): y = E0 + δ log(x) E0 … intercept, δ … slope• ExponentialExponential decay („drm”): y = c + (d - c)(exp(-x/e)) c,d … upper and lower limit, e …steepnes of decay/growth Exponential Model ((“fitMod”): y = E0 + E1(exp(x/δ) - 1) E0 … placebo effect, E1 … slope, δ … Parameter, controlling the convexity of the model• PeakGaussian („drm”): y = c+(d-c)*exp(-0.5 * ((x-e)/b)^2) c … background effect, d... peak, e … critical point, b … growth rateLog-gaussian („drm”): y = c+(d-c)*exp(-0.5 * ((log(dose)-log(e))/b)^2) c … background effect, d ... peak, e … critical point, b … growth rate5P Gaussian („drm”): y = c+(d-c)*exp(-0.5 * ((x-e)/b)^f) c … background effect, d... peak, e … critical point, b … growth rate5P Log-gaussian („drm”): y = c+(d-c)*exp(-0.5 * (sqrt(((log(x)-log(e))/b)^2))^f) c … background effect, d ... peak, e … critical point, b … growth rate• Sigmoidal4P Logistic (“fitMod”): y = E0 + Emax / (1 + exp ((ED50 - x) / δ)) E0 … left-asymptote corresponding to a basal effect level, Emax … Asymptotic maximum effect, δ … steepness of the curve4P Log-Logistic; Hill (“fitMod”): y = E0 + Emax (x^h / (ED50 ^h +x^h)) E0 … left-asymptote corresponding to a basal effect level, Emax … Asymptotic maximum effect, h … steepness of the curve (Hill parameter)Gompertz (“drc”) : y = c + (d - c)(exp(-exp(b(x-e)))) Weibull type I („drc“): y = c + (d - c)exp(-exp(b(log(x)-log(e))))Michaelis-Menten (“fitMod”): y = E0 + Emax (x / (ED50 +x)) E0 … placebo effect, Emax … Asymptotic maximum effect To change symbols in plots selectsymbol numbers from e.g.http://www.endmemo.com/program/R/pchsymbols.php Concentration-response modelling single or multiple replicatesusing PrecistionTox templatesigmoidal, log-logistic, Hill-slope modelConstraints for min and max are set to 0 and 100. The loop design allows to group files with replicates of the same chemical and toanalyse several endpoints simultanously. Note that the workflow may need adaptation in case the file name structure and data tablestructure is changing.stefan.scholz(at)ufz.de Node 696Save a temporary fileRemove columnsnot required for modelingAdjust alsoempty table configurationto align with output ofmodel nodeSelect symbolsNode 974Node 977Node 978Modelled curvesNode 1075Node 1114Node 1115Node 1195Node 1196Node 1198Include "yes"Convert concentration, replicateand effect to number (optional)Convert INTto DOUBLEExtract experiment IDNode 1220Node 1222Node 1224Definitionof parametersDeleteempty rowsSelect also Hillparameters Raw dataNode 1236To replace group loop startfor improving speedNode 1255Node 1256Node 1257Node 1258ReportingNode 1260Node 1269Node 1273Endpoint explanationNode 1276Node 1277Node 1278Node 1282Node 1283Node 1290Node 1291Node 1293optional row filterfor testingspecific data setsCalculate effect threshold(min+10)Node 1298Node 1299Node 1300Node 1301Node 1302Node 1303 Collect metainformation Excel Writer (XLS)(deprecated) Donwload Excelresult file Column Filter Conc-resp modelling Table Creator Table Creator Catch Errors(Data Ports) Try (Data Ports) XLS Sheet Appender(deprecated) Create TempDir (legacy) Number To String String Replacer Math Formula String To Number Column AutoType Cast View results Row Filter String To Number Column Rename GroupBy Joiner (deprecated) Missing Value Column Resorter XLS Sheet Appender(deprecated) Row Filter File upload andparameter selection XLS Sheet Appender(deprecated) GroupBy Grouping Remove values fromselected experiments View results Joiner (deprecated) Row Filter Prepare model parametersto join with raw data Sorter Excel Sheet Appender(XLS) (deprecated) Missing Value Column Resorter Arrange images XLS Sheet Appender(deprecated) Column Rename Column AutoType Cast Column Resorter Inspectuploaded data Joiner Table Rowto Variable Chunk Loop Start Loop End Math Formula Automatically flagand remove BMDs Row Filter Math Formula Table Validator CASE Switch End Excel Reader Excel Reader Excel Reader String To Number

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: sscholz
Created at: 2016-01-08
On NodePit since: 2026-09-17
Last update: 2026-09-17
Created with KNIME version: v4.7.8
Tags: Range finding

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!