plotLinear: Plot linear model for sequins

In Anaquin: Statistical analysis of sequins

Description

Create linear model for sequins, between input concentation on the x-axis and measurment on the y-axis.

Usage

plotLinear(seqs, x, y, std, title, xlab, ylab, showSD, showLOQ, showStats,
           xBreaks, yBreaks, errors, showLinear, showAxis)

Arguments

seqs	Sequin names
x	Input concentration on the x-axis
y	Measurement on the y-axis
std	Standard deviation. (Default to NULL).
title	Label of the plot. (Default to NULL).
xlab	Label for the x-axis. (Default to NULL).
ylab	Label for the y-axis. (Default to NULL).
xBreaks	Breaks for the x-axis. (Default to NULL).
yBreaks	Breaks for the y-axis. (Default to NULL).
showSD	Display vertical standard deviation bars. (Default to FALSE).
showLOQ	Display limit-of-quantification? Default to TRUE.
showStats	Display regression statistics? Default to TRUE.
errors	How errors bar should be calculated. SD or Range.
showLinear	Display regression line. (Default to TRUE).
showAxis	Display x-axis and y-axis. (Default to TRUE).

Details

The plotLinear function plots a scatter plot with input concentration on the x-axis, and measurement on the y-axis. The input concentration is typically the concentration level in ladder mixture, although other measures (such as expected copy number) are also possible. The function builds a linear regression between the two variables, and reports associated statistics (R2, correlation and regression parameters) on the plot.

The function also estimates limit-of-quantification (LOQ) breakpoint, and reports it on the plot if found. LOQ is defined as the lowest empirical detection limit, a threshold value beyond which stochastic behavior occur. LOQ is estimated by fitting segmented linear regression with two segments on the entire data set, while minimizing the total sum of squares of the differences between the variables.

Value

The function prints a scatter plot and return it's LOQ statistics.

Author(s)

Ted Wong [e-mail]

Examples

library(Anaquin)

#
# Data set generated by Cufflinks and Anaquin. described in Section 5.4.6.3 of
# the user guide.
#
data(UserGuideData_5.4.6.3)

title <- 'Gene Expression'
xlab  <- 'Input Concentration (log2)'
ylab  <- 'FPKM (log2)'

# Sequin names
seqs <- row.names(UserGuideData_5.4.6.3)

# Input concentration
x <- log2(UserGuideData_5.4.6.3$Input)

# Measured FPKM
y <- log2(UserGuideData_5.4.6.3[,2:4])

plotLinear(seqs, x, y, title=title, xlab=xlab, ylab=ylab, showLOQ=TRUE)

Example output

Loading required package: ggplot2
$LOQ
NULL

Warning messages:
1: Use of `data$x` is discouraged. Use `x` instead. 
2: Use of `data$y` is discouraged. Use `y` instead. 
3: Use of `data$x` is discouraged. Use `x` instead. 
4: Use of `data$y` is discouraged. Use `y` instead. 
5: Use of `data$x` is discouraged. Use `x` instead. 
6: Use of `data$ymin` is discouraged. Use `ymin` instead. 
7: Use of `data$x` is discouraged. Use `x` instead. 
8: Use of `data$ymax` is discouraged. Use `ymax` instead. 

Anaquin documentation built on Nov. 8, 2020, 5:23 p.m.