Create Limit-of-Detection Ratio (LODR) plot

Share:

Description

Create a Limit-of-Detection Ratio (LODR) plot between measured abundance (x-axis) and p-value probability (y-axis).

Usage

1
plotLODR(data, FDR, title, xlab, ylab, legTitle, showConf, ...)

Arguments

data

Anaquin dataset created by AnaquinData. It needs to define information in Details.

FDR

Chosen false-discovery-rate. 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.

legTitle

Title for the legend. Default to 'Ratio'.

showConf

Show confidence interval? Default to FALSE.

...

Reserved for internal testing

Details

plotLODR requires the following data inputs from AnaquinData.

seqs List of sequin identifiers (eg. R2_11_2).
measured Measured abundance (eg: average counts, DP field in a VCF file etc)
ratio Expected ratio; eg: expected log-fold ratio or expected allele frequency etc
pval P-value probability

Create a Limit-of-Detection Ratio (LODR) plot between measured abundance (x-axis) and p-value probability (y-axis).

The LODR plot indicates the confidence in measurement relative to the magnitude of the measurement. For example, p-value should converge to zero as the sequencing depth increases.

The function also fits non-parametric curves for each sequin ratio group. The curves are modelled with local regression analysis, and are colored by the sequin group.

plotLODR is an amendment from the LODR code in the ERCC dashboard R-package. Further details on the statistical algorithm is available in the ERCC documentation at https://bioconductor.org/packages/release/bioc/html/erccdashboard.html.

Value

The functions does not return anything but it prints a LODR plot.

Author(s)

Ted Wong t.wong@garvan.org.au

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
library(Anaquin)

#
# Data set generated by DESeq2 and Anaquin. described in Section 5.6.3.3 of
# the user guide.
#
data(UserGuideData_5.6.3)

xlab  <- 'Average Counts'
ylab  <- 'P-value'
title <- 'LODR Curves'

# Sequin names
seqs <- row.names(UserGuideData_5.6.3)

# Expected log-fold
ratio <- UserGuideData_5.6.3$ExpLFC

# Measured average abundance
measured <- UserGuideData_5.6.3$Mean

# P-value
pval <- UserGuideData_5.6.3$Pval

# Q-value
qval <- UserGuideData_5.6.3$Qval

anaquin <- AnaquinData(analysis='PlotLODR',
                           seqs=seqs,
                      measured=measured,
                         ratio=ratio,
                          pval=pval,
                          qval=qval)

plotLODR(anaquin, xlab=xlab, ylab=ylab, title=title, FDR=0.1)