imd_test: Tests for the independence of missing data across groups (aka...

In pmartR/pmartR: Panomics Marketplace - Quality Control and Statistical Analysis for Panomics Data

Tests for the independence of missing data across groups (aka factors, aka main effects)

Description

Tests the null hypothesis that the number of missing observations is independent of the groups. A g-test is used to test this null hypothese against the alternative that the groups and missing data are related. This is usually performed in conjuction with an ANOVA which tests if the mean response (which varies with data type) is the same across groups; this combination is called IMD_ANOVA. It's probably a good idea to first filter the data with 'imd_anova_filter' to see if there is enough infomration to even do this test. See Webb-Robertson et al. (2010) for more.

Usage

imd_test(
  omicsData,
  groupData,
  comparisons,
  pval_adjust_multcomp,
  pval_adjust_fdr,
  pval_thresh,
  covariates,
  paired,
  parallel = TRUE
)

Arguments

omicsData	A pmartR data object of any class
groupData	'data.frame' that assigns sample names to groups
comparisons	'data.frame' with columns for "Control" and "Test" containing the different comparisons of interest. Comparisons will be made between the Test and the corresponding Control If left NULL, then all pairwise comparisons are executed.
pval_adjust_multcomp	A character string specifying the type of multiple comparisons adjustment to implement. The default setting, "none", is to not apply an adjustment. Valid options include: "bonferroni" and "holm".
pval_adjust_fdr	A character string specifying the type of FDR adjustment to implement. The default setting, "none", is to not apply an adjustment. Valid options include: "bonferroni", "BH", "BY", and "fdr".
pval_thresh	numeric p-value threshold, below or equal to which peptides are considered differentially expressed. Defaults to 0.05
covariates	A character vector with no more than two variable names that will be used as covariates in the IMD-ANOVA analysis.
paired	A logical value that determines whether paired data should be accounted for
parallel	A logical value indicating whether or not to use a "doParallel" loop when running the G-Test with covariates. The default is TRUE.

Value

a list of 'data.frame's

Results	e_data cname, Count of non-missing data for each group, Global G-test statistic and p-value
Gstats	Value of the g statistics for each of the pairwise comparisons specified by the `comparisons` argument
Pvalues	p-values for each of the pairwise comparisons specified by `comparisons` argument
Flags	Indicator of statistical significance where the sign of the flag reflects the difference in the ratio of non-missing observations (0/+-2 to if adjusted p-value>=pval_thresh or p-value<pval_thresh)

Author(s)

Bryan Stanfill

References

Webb-Robertson, Bobbie-Jo M., et al. "Combined statistical analyses of peptide intensities and peptide occurrences improves identification of significant peptides from MS-based proteomics data." Journal of proteome research 9.11 (2010): 5748-5756.

pmartR/pmartR documentation built on March 4, 2024, 8:32 a.m.