README.md
In HHayman/FlowR: Carries out a selected workflow

FlowR

FlowR consists of a series of workflows, each representing a form of analysis. To date, these are; CutFlow, SurvFlow and CompFlow.

You can install the released version of FlowR from GITHUB with:

library(devtools)
install_github("HHayman/FlowR")

You can load the FlowR package with:

library(FlowR)

Arrange your data to meet the following criteria;

Carry out any exclusions
Include an identifier variable
Your status variable needs to be coded as 0 and 1, 0 for no event, 1 for an event
Time variable must be included as a a continuous variable
There is no limit on the number of variables you wish to generate a cutpoint for
If coding multiple datasets, the respective variables must have the exact same name in all datasets
- Additional datasets do not need to contain all variables from your training dataset, just those you wish to be coded in that dataset
Save your dataset(s) as a CSV file
Create a subdirectory in your R directory, and place your dataset files inside

To run CutFlow, simply fill in the blanks, as in the example below;

Subdirectory is the name of the folder you placed your datasets in. It must be within your current directory
TrainingData is the name of the dataset you wish to be used to generate the cutpoint
- This cutpoint is then applied to all datasets within the subdirectory
CutPointStatus is the status variable to be used (such as CSS)
CutPointTime is the time variable to be used (such as Survival_2021)
minprop is the minimum proportion of cases to be include either side of the cutpoint
- Default is 0.1, exclude the argument if you don’t want to change this
Greyscale is an optional toggle to produce a greyscale variant of all plots
- The default is colour, exclude the argument if you don’t want to change this
Variables is a list of your variables to generate cutpoints for

CutFlow(Subdirectory = "YourSubdirectory", TrainingData = "TrainingDataset", 
        CutPointStatus = "StatusVariable", CutPointTime = "TimeVariable", minprop = 0.1, 
        Greyscale = TRUE, Variables = c( "Variable1", 
                                         "Variable2",  
                                         "Variable3", 
                                         "Variable4",
                                         "Variable5", 
                                         "Variable6",  
                                         "Variable7", 
                                         "Variable8",
                                         "Variable9", 
                                         "Variable10",  
                                         "Variable11", 
                                         "Variable12"))

A new folder will be created in your R directory;

Folder name format is CutFlow_SystemData_Number
Three folders are contained within;
- 0 - A copy of all datafiles fed into CutFlow, for record keeping
- 1 - A copy of all cutpoint data, including a pdf list of cutpoints
- 2 - A copy of all datasets, newly coded

SurvFlow consists of modules. Currently there are two, SurvBase and SurvFacet

SurvBase generates KM plots per variable, and provides x-year survival, logrank, hazard ratio, case numbers and event numbers
SurvFacet runs SurvBase per level of each provided facet, such as TNM stage or MSI status
SurvFacet is currently disabled but will be enabled again in the near future

Arrange your data to meet the following criteria;

Carry out any exclusions
Include an identifier variable
Your status variables needs to be coded as 0 and 1, 0 for no event, 1 for an event
Time variables must be included as a a continuous variable
There is no limit on the number of variables you wish to generate survival output for
Save your dataset as a CSV file

To run SurvFlow, simply fill in the blanks, as in the example below;

Data is a Coded dataset in csv format. If using CutFlow, simply use the produced coded dataset
Variables is a list of variables (coded as 0 and 1) for analysis
LegendLabels are optional labels for your legends. Default is the value of the level (0 and 1)
Identifier is an identifier variable for cases
PlotTitles are optional plot titles. Default is variable name
SurvivalStatus are status variables (coded as 0 and 1)
- Must have the same number of elements as the SurvivalTime variable
SurvivalTime are survival time variables - continuous
- Must have the same number of elements as the SurvivalStatus variable
SurvivalTimeUnit is the unit of time for survival time
xYearSurvivalVar is the number of years to be used to calculate ‘X’ years survival. Default = 5
SurvBase takes TRUE or FALSE. Toggle to activate base survival analysis.

Data <- read.csv(file.choose(), fileEncoding = 'UTF-8-BOM')


SurvFlow(
  Data,
  Variables = c(
    "Variable1",
    "Variable2",
    "Variable3",
    "Variable4",
    "Variable5",
    "Variable6",
    "Variable7",
    "Variable8",
    "Variable9",
    "Variable10",
    "Variable11",
    "Variable12"
  ),
  LegendLabels = c("Low", "High"),
  Identifier = "Identifier",
  PlotTitles = c(
    "Title1",
    "Title2",
    "Title3",
    "Title4",
    "Title5",
    "Title6",
    "Title7",
    "Title8",
    "Title9",
    "Title10",
    "Title11",
    "Title12"
  ),
  SurvivalStatus = c("Status1", "Status2", "Status3", "Status4"),
  SurvivalTime = c("Time1", "Time1", "Time2", "Time2"),
  xYearSurvivalVar = 5,
  SurvivalTimeUnit = "Months",
  SurvBase = TRUE
)

A new folder will be created in your R directory;

Folder name format is SurvFlow_Filename_SystemData_Number
Inside is a folder per SurvFlow module, for example BaseSurv
- At the next level is a folder per survival status/time pair, containing the survival plots
The plot can be seen as below;

CompFlow compares two sets of measurements and outputs;

Scatterplot
Bland-Altman plot
% of points on the Bland-Altman which fall outside of confidence limts
ICCC value
A quantification of proportional bias (Pearson) - only if your data meet assumptions for Pearson

Arrange your data to meet the following criteria;

2 columns of equal length

To run CompFlow, simply fill in the blanks, as in the example below;

Data is a CSV file containing your two columns of data
ICCModel - must be ‘oneway’ or ‘twoway’. See ‘(A) “Model” Selection’ in ICC reference
ICCTYpe - must be ‘consistency’ or ‘agreement’. See ‘(C) “Definition” Selection’ in ICC reference
ICCUnit - must be ‘single’ or ‘average’. See ‘(B) “Type” Selection’ in ICC reference
Density - must be ‘Yes’, ‘No’. Density is ‘No’ by default. This adds density onto your plots.

ICC reference

Data <- read.csv(file.choose(), fileEncoding = 'UTF-8-BOM')


CompFlow(Data, ICCModel = "twoway", ICCType = "agreement", ICCUnit = "single", Density = "Yes")

A new folder will be created in your R directory;

Folder name format is CompFlow_SystemData_Number
Inside are three folders
- 0.OriginalData contains your input file for record keeping
- 1.ConfiguredData contains records of data configurations for record keeping
- 2.Plots contains your output
The plots can be seen as below;