control_adjustment: Calculates a corrected control group, discovers outliers in...

In jeitziner/TTMap: Two-Tier Mapper: a clustering tool based on topological data analysis

Description

control_adjustment function finds outliers in the control group and removes them

Usage

    control_adjustment(normal.pcl, tumor.pcl, normalname, dataname, 
    org.directory = "", A = 1, e = 0, meth = 0, P = 1.1, B = 0)

Arguments

normal.pcl	the control matrix with annotation as obtained by $CTRL from make_matrices
tumor.pcl	the disease/test data matrix with annotation as obtained by $TEST from make_matrices
normalname	A name for the corrected control files
dataname	the name of the project
org.directory	where the outputs should be saved
A	integer if A=0 then the difference to the median is calculated otherwise the difference to the mean.
e	integer giving how far to the median an outlier is at least
meth	value or method that defines how to replace outliers, default is set to replace by the median
P	if more than P percent of features are outliers the feature is removed, by default all are kept
B	Batch vector a vector for normal and test samples with a same number corresponding to a same batch

Details

control_adjustment calculates a corrected control group, discovers outliers in it.

Value

Several files are created

paste(org.directory,normalname,".normMesh",sep = "")	The normal matrix with only common features with the test matrix. This file is only created if the two have different rows
paste(org.directory,dataname,".normMesh",sep = "")	The test matrix with only common features with the normal matrix. This file is only created if the two have different rows.
mean_vs_variance.pdf	A pdf showing a plot of the mean (X axis) against the variances (Y axis) of each feature
mean_vs_variance_after_correction.pdf	A pdf showing a plot of the mean (X axis) against the variances (Y axis) of each feature after correction of the control group
na_numbers_per_row.txt	number of outliers per row
na_numbers_per_col.txt	number of outliers per column

And values of ttmap_part1_ctrl_adj

e	Selected criteria for what is an outlier
tag.pcl	Annotation of features, ID of features and weight
Normal.mat	The control matrix without annotation and only with the common rows with Disease.mat
Disease.mat	The test/disease matrix without annotation and only with the common rows with Disease.mat
flat.Nmat	A list $mat being the corrected control matrix $m a record of the different numbers of removed genes per sample
record	numbers recording the number of columns in Disease.mat and Normal.mat
B	The batch vector B introduced in the begining
U1	The different batches in Normal.mat
U2	The different batches in Disease.mat

Author(s)

Rachel Jeitziner

See Also

hyperrectangle_deviation_assessment, ttmap ttmap_sgn_genes

Examples

    ##--
    library(airway)
    data(airway)
    airway <- airway[rowSums(assay(airway))>80,]
    assay(airway) <- log(assay(airway)+1,2)
    ALPHA <- 1
    the_experiment <- TTMap::make_matrices(airway,
    seq_len(4), seq_len(4) + 4,
    rownames(airway), rownames(airway))
    TTMAP_part1prime <-TTMap::control_adjustment(
    normal.pcl = the_experiment$CTRL,
    tumor.pcl = the_experiment$TEST, 
    normalname = "The_healthy_controls", 
    dataname = "Effect_of_cancer", 
    org.directory = tempdir(), e = 0, P = 1.1, B = 0);

jeitziner/TTMap documentation built on May 23, 2019, 4:24 p.m.