xCorrelation: Function to calculate and visualise correlation

In Pi: Leveraging Genetic Evidence to Prioritise Drug Targets at the Gene and Pathway Level

Description

xCorrelation is supposed to calculate and visualise correlation between a data frame and a named vector (or a list of named vectors).

Usage

xCorrelation(
df,
list_vec,
method = c("pearson", "spearman"),
p.type = c("nominal", "empirical"),
seed = 825,
nperm = 2000,
p.adjust.method = c("BH", "BY", "bonferroni", "holm", "hochberg",
"hommel"),
plot = FALSE,
plot.smooth = c(NA, "lm", "loess")
)

Arguments

df	a data frame with two columns ('name' and 'value')
list_vec	a named vector containing numeric values. Alternatively it can be a list of named vectors
method	the method used to calcualte correlation. It can be 'pearson' for Pearson's correlation or 'spearman' for Spearman rank correlation
p.type	the type of the p-value calcualted. It can be 'nominal' for nominal p-value or 'empirical' for empirical p-value
seed	an integer specifying the seed
nperm	the number of random permutations
p.adjust.method	the method used to adjust p-values. It can be one of "BH", "BY", "bonferroni", "holm", "hochberg" and "hommel". The first two methods "BH" (widely used) and "BY" control the false discovery rate (FDR: the expected proportion of false discoveries amongst the rejected hypotheses); the last four methods "bonferroni", "holm", "hochberg" and "hommel" are designed to give strong control of the family-wise error rate (FWER). Notes: FDR is a less stringent condition than FWER
plot	logical to indicate whether scatter plot is drawn
plot.smooth	the smooth method for the scatter plot. It can be NA (depending on correlation type), "lm" for the linear line or 'loess' for the loess curve

Value

a list with three componets:

• df_summary: a data frame of n x 5, where n is the number of named vectors, and the 5 columns are "name", "num" (i.e. number of data points used for calculation), "cor" (i.e. correlation), "pval" (i.e. p-value), "fdr"

• ls_gp_curve: NULL if the plot is not drawn; otherwise, a list of 'ggplot' objects for scatter plot together with an estimated curve

• ls_gp_pdf: NULL if the plot is not drawn; otherwise, a list of 'ggplot' objects for pdf plot for null distribution of correlation together with a vertical line for observed correlation

Note

none

See Also

xCorrelation

Examples

RData.location <- "http://galahad.well.ox.ac.uk/bigdata"
## Not run: 
# a) provide the seed nodes/genes with the weight info
## load ImmunoBase
ImmunoBase <- xRDataLoader(RData.customised='ImmunoBase',
RData.location=RData.location)
## get genes within 500kb away from AS GWAS lead SNPs
seeds.genes <- ImmunoBase$AS$genes_variants
## seeds weighted according to distance away from lead SNPs
data <- 1- seeds.genes/500000

# b) prepare a data frame
df <- data.frame(name=names(data), value=data, stringsAsFactors=FALSE)

# c) do correlation
ls_res <- xCorrelation(df, data, method="pearson", p.type="empirical",
nperm=2000, plot=TRUE)

## End(Not run)

Pi documentation built on Nov. 26, 2020, 2:01 a.m.