peptide_cooccurrence: Summarize the co-occurrence of two peptides

Description Usage Arguments Value Author(s) See Also Examples

View source: R/peptide_cooccurrence.R

Description

calculate a contingency table of two peptides' presence in multiple individuals and perform independence test and cooccurrence coefficients

Usage

1
2
3
4
5
6
7
peptide_cooccurrence(
  x,
  y,
  coefficient = "jaccard",
  perform_test = FALSE,
  test_method = "fisher"
)

Arguments

x

a logical vector of peptide 1's presence in multiple samples

y

a logical vector of peptide 2's presence in multiple samples

coefficient

the cooccurrence coefficient to compute. Options: "jaccard" (jaccard index), "phi" (phi correlation coefficient), and "prop" (mean proportion of both peptide being enriched given one of them is enriched), or a vector of any combination of the three options (e.g., c("jaccard", "phi")). Default: "jaccard"

perform_test

a logical indicator of whether statistical tests should be performed for each comparison. The test method is specified by test_method. Default: FALSE

test_method

method to perform statistical test. Options: "fisher" (Fisher's exact test) or "chisq" (Pearson's Chi-squared test). Default: "fisher"

Value

if coefficient = "phi" and perform_test = FALSE, return a single number, i.e., the phi coefficient. Otherwise, return a vector that contains at least these four elements: TT: frequency of both peptide positive; TF: frequency of peptide 1 positive and peptide 2 negative; FT: frequency of peptide 1 negative and peptide 2 position; FF: frequency of both peptide negative. Depending on coefficient, the vector also contains jaccard index, phi coefficient and/or the mean proportion of cooccurrence. If perform_test is TRUE, the vector also contains the key statistics ("oddsr" - odds ratio, if using the Fisher's exact test, or "xsq" - chi-squared, if using the Chi-squared test), and the p-value.

Author(s)

Siyang Xia sxia@hsph.harvard.edu

See Also

peptide_pairwise_correlation, fisher.test, chisq.test

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
data(peptide_z)
peptide_bi <- as.data.frame((peptide_z > 3.5) * 1)
peptide_cooccurrence(x = peptide_bi[, 1], y = peptide_bi[, 2],
                     coefficient = "jaccard",
                     perform_test = TRUE,
                     test_method = "fisher")
peptide_cooccurrence(x = peptide_bi[, 1], y = peptide_bi[, 2],
                     coefficient = "jaccard",
                     perform_test = TRUE,
                     test_method = "chisq")
peptide_cooccurrence(x = peptide_bi[, 1], y = peptide_bi[, 2],
                     coefficient = c("phi", "prop", "jaccard"),
                     perform_test = FALSE)

siyangxia419/virlink documentation built on Jan. 2, 2022, 12:16 a.m.