Skip to content

Command Line

Jump to bottom
japhill edited this page Jan 17, 2021 · 29 revisions

Overview

The BMDExpress 2 Command Line gives access to the full workflow of the BMDExpress application without having to navigate through a GUI. The results produced from the command line version can then be easily imported back into the application if the need arises. The command line can be run in 4 different modes: analyze, query, export, and delete.

Video Tutorials

Accessing the command line version

Analyze data with command line version

Export, delete and query

Analyze

Example Configuration Files

bmdexpress2-cmd analyze --config-file

--config-file

The configuration file has four main configuration sections which are expressionDataConfigs, preFilterConfigs, bmdsConfigs, and categoryAnalysisConfigs.

The expressionDataConfigs section is used to detail the initial input file containing the expression data. It contains fields to describe the platform the data comes from as well as whether or not the data was log transformed or not. Each instance of this section will create a single expression data output in the final bm2 file.

The next section is the preFilterConfigs section. This section is used to describe the different prefilters that should be used on the initial expression data before running the BMD Analysis. There is a choice between three different prefilters which are anova, williams, and oriogen. Each instance of this section will create a single prefilter output corresponding to the type of prefilter chosen.

The next section is the bmdsConfigs which outlines the details for the BMDAnalysis. This section has three different subsections which are the modelConfigs, bmdsBestModelSelection, and bmdsInputConfigs. The modelConfigs subsection defines the models to run. The bmdsBestModelSelection subsection defines the parameters that are used to determine how the best model is selected. The bmdsInputConfigs subsection defines input parameters that are fed to the individual models during execution. Each instance of the overall bmdsConfigs section will generate a Benchmark Dose Analysis output in the bm2 file.

The last section is the categoryAnalysisConfigs which is used to define any category analyses you would like to compute on any previously generated BMD Analysis. There is a choice between three different category analysis types which are defined, go, and pathway. Each instance of this section will take a specified BMD Analysis and generate a single Functional Classification output in the bm2 file.

The full list of fields for each section can be found below:

(* indicate required fields)

  • bm2FileName - The file name of the bm2 file that will be outputted
  • jsonExportFileName - The file name of the json export file
  • overwrite - Choice between (true | false)
  • basePath – override the default base path for storing meta data and annotation files for bmdexpress2. The default location is in the user home directory under "bmdexpress2." This option allows localization of file input/output for running many instances of the application in massive parallel on many compute nodes.
  • expressionDataConfigs
    • inputFileName* - The file or directory with files to import
    • outputName - If not specified, the imported filename will be used
    • platform* - The name of the platform (Use ID column from annotations table)
    • logTransformation* - Choice between (NONE | BASE2 | BASE10 | NATURAL)
  • preFilterConfigs
    • @type* - Choice between (anova | williams | oriogen)
    • inputName - If not specified, runs filter on all expression data sets
    • outputName - If not specified, assigns a default name
    • pValueCutoff* - Numeric value between 0 - 1
    • useMultipleTestingCorrection* - Choice between (true | false)
    • filterOutControlGenes* - Choice between (true | false)
    • useFoldChange* - Choice between (true | false)
    • foldChange* - Numeric value
    • foldChangeLotel - Numeric value
    • pValueLotel - Numeric value
    • lotelTest - Choice between (1 | 2)
      • 1 - Dunnett's Test
      • 2 - T-Test
    • numberOfThreads - Numeric value
    • numberOfPermutations* - Integer value [Only for Williams]
    • initialBootstraps* - Integer value [Only for Oriogen]
    • maxBootstraps* - Integer value [Only for Oriogen]
    • s0Adjustment* - Numeric value [Only for Oriogen]
  • bmdsConfigs
    • modelConfigs*
      • @type* - Choice between (poly | power | exp | hill)
      • expModel* - Choice between (2 | 3 | 4 | 5) [Only for exponential]
      • degree* - Choice between (1 | 2 | 3 | 4) [Only for poly]
    • bmdsBestModelSelection*
      • bestPolyTest* - Choice between (1 | 2)
        • 1 - Nested Chi Square
        • 2 - Lowest AIC
      • pValueCutOff* - Numeric value
      • flagHillWithKParameter* - Choice between (true | false)
      • kParameterValue* - Choice between (1 | 2 | 3)
        • 1 - Lowest Positive Dose
        • 2 - 1/2 of Lowest Positive Dose
        • 3 - 1/3 of Lowest Positive Dose
      • bestModelSelectionWithFlaggedHill* - Choice between (1 | 2 | 3 | 4 | 5)
        • 1 - Include Flagged Hill
        • 2 - Exclude Flagged Hill from Best Models
        • 3 - Exclude All Hill from Best Models
        • 4 - Modify BMD if Flagged Hill as Best Model
        • 5 - Select Next Best Model with P-Value > 0.05
      • modifyFlaggedHillWtihFractionMinBMD* - Numeric value
    • bmdsInputConfig*
      • maxIterations* - Integer value
      • confidenceLevel* - Numeric value
      • constantVariance - Choice between (true | false)
      • restrictPower - Choice between (true | false)
      • bmrFactor* - Numeric value
    • inputCategory* - Choice between (anova | williams | oriogen | expression)
    • inputName - If not specified, runs all “inputCategory” datasets
    • outputName - If not specified, outputs default name
    • killTime – number of seconds to allow models to run before destroying them. (default 600)
    • tmpFolder – override the temp folder used to read and write bmds model files. The specified folder also needs to allow execution of programs because it will also store a copy of the model executables. This option helps when running massive parallel instances of the software and greatly reduces file input/output to a home directory.
    • numberOfThreads* - Integer value
  • categoryAnalysisConfigs
    • @type* - Choice between (gene | defined | go | pathway)
    • inputName - If not specified, then all bmd analysis is run
    • outputName - If not specified, then default name is assigned
    • removePromiscuousProbes* - Choice between (true | false)
    • removeBMDGreaterHighDose* - Choice between (true | false)
    • identifyConflictingProbeSets* - Choice between (true | false)
    • correlationCutoffForConflictingProbeSets* - Numeric value
    • bmdPValueCutoff - Numeric value
    • bmdBMDLRatioMin - Numeric value
    • bmduBMDRatioMin - Numeric value
    • bmduBMDLRatioMin - Numeric value
    • nFoldBelowLowestDose - Numeric value
    • maxFoldChange - Numeric value
    • prefilterPValueMin - Numeric value
    • probeFilepath* - File path to probe to gene mapping [Only for defined]
    • categoryFilepath* - File path to the gene to category mapping [Only for defined]
    • deduplicateGeneSets - Choice between (true | false)
    • goCategory* - Choice between (universal | biological_process | molecular_function | cellular components) [Only for GO]
    • signalingPathway* - Choice between (REACTOME | BIOPLANET) [Only for pathway]

Query

The query command lets you print the names of all analyses from a given group out to the command line. This command takes two inputs, analysis-group, input-bm2 with the following form:

bmdexpress2-cmd query --analysis-group <GROUP> --input-bm2 <BM2FILE>

input-bm2 : This is the bm2 file that contains the project you are working on

analysis-group: Choice from (expression | anova | williams | oriogen | bmd | categorical)

Export

The export command lets you export specific analyses from a given project out to a file. This command takes three inputs, analysis-group, input-bm2, output-file, with an optional fourth, analysis-name. If the analysis-name is left out, then all the analyses in the group are exported into a single file. The specific command has the following form:

bmdexpress2-cmd export --analysis-group <GROUP> [--analysis-name <NAME>] --input-bm2 <BM2FILE> --output-file-name <OUTPUT>

input-bm2 : This is the bm2 file that contains the project you are working on

analysis-group: Choice from (expression | anova | williams | oriogen | bmd | categorical)

analysis-name: The name of the specific analysis that you want to delete

output-file-name: The name of the file that will be output

Delete

The delete command lets you delete specific analyses from a given project. This command takes three inputs, analysis-group, analysis-name, and input-bm2 in the following form:

bmdexpress2-cmd delete --analysis-group <GROUP> --analysis-name <NAME> --input-bm2 <BM2FILE>

input-bm2 : This is the bm2 file that contains the project you are working on

analysis-group: Choice from (expression | anova | williams | oriogen | bmd | categorical)

analysis-name: The name of the specific analysis that you want to delete

Clone this wiki locally