Package 'HIMA' reference manual

Title:	High-Dimensional Mediation Analysis
Description:	Allows to estimate and test high-dimensional mediation effects based on advanced mediator screening and penalized regression techniques. Methods used in the package refer to Zhang H, Zheng Y, Zhang Z, Gao T, Joyce B, Yoon G, Zhang W, Schwartz J, Just A, Colicino E, Vokonas P, Zhao L, Lv J, Baccarelli A, Hou L, Liu L. Estimating and Testing High-dimensional Mediation Effects in Epigenetic Studies. Bioinformatics. (2016) <doi:10.1093/bioinformatics/btw351>. PMID: 27357171.
Authors:	Yinan Zheng [aut, cre] , Haixiang Zhang [aut], Lifang Hou [aut], Lei Liu [aut, cph]
Maintainer:	Yinan Zheng <[email protected]>
License:	GPL-3
Version:	2.3.0
Built:	2024-10-23 06:20:17 UTC
Source:	https://github.com/YinanZheng/HIMA

High-Dimensional Mediation Analysis for 'Omic' Data

Description

HIMA is an R package for estimating and testing high-dimensional mediation effects in omic studies. HIMA can perform high-dimensional mediation analysis on a wide range of omic data types as potential mediators, including epigenetics, transcriptomics, proteomics, metabolomics, and microbiomics. HIMA can also handle survival data mediation analysis and perform quantile mediation analysis.

Package:	HIMA
Type:	Package
Version:	2.3.0
Date:	2024-10-21
License:	GPL-3

Details

# If package "qvalue" is not found during installation, please first install "qvalue" package # through Bioconductor: https://www.bioconductor.org/packages/release/bioc/html/qvalue.html

Author(s)

Yinan Zheng [email protected], Haixiang Zhang [email protected], Lei liu (Contact) [email protected]

Maintainer: Yinan Zheng [email protected]

References

1. Zhang H, Zheng Y, Zhang Z, Gao T, Joyce B, Yoon G, Zhang W, Schwartz J, Just A, Colicino E, Vokonas P, Zhao L, Lv J, Baccarelli A, Hou L, Liu L. Estimating and Testing High-dimensional Mediation Effects in Epigenetic Studies. Bioinformatics. 2016. DOI: 10.1093/bioinformatics/btw351. PMID: 27357171; PMCID: PMC5048064

2. Zhang H, Zheng Y, Hou L, Zheng C, Liu L. Mediation Analysis for Survival Data with High-Dimensional Mediators. Bioinformatics. 2021. DOI: 10.1093/bioinformatics/btab564. PMID: 34343267; PMCID: PMC8570823

3. Zhang H, Chen J, Feng Y, Wang C, Li H, Liu L. Mediation Effect Selection in High-dimensional and Compositional Microbiome data. Stat Med. 2021. DOI: 10.1002/sim.8808. PMID: 33205470; PMCID: PMC7855955

4. Zhang H, Chen J, Li Z, Liu L. Testing for Mediation Effect with Application to Human Microbiome Data. Stat Biosci. 2021. DOI: 10.1007/s12561-019-09253-3. PMID: 34093887; PMCID: PMC8177450

5. Perera C, Zhang H, Zheng Y, Hou L, Qu A, Zheng C, Xie K, Liu L. HIMA2: High-dimensional Mediation Analysis and Its Application in Epigenome-wide DNA Methylation Data. BMC Bioinformatics. 2022. DOI: 10.1186/s12859-022-04748-1. PMID: 35879655; PMCID: PMC9310002

6. Zhang H, Hong X, Zheng Y, Hou L, Zheng C, Wang X, Liu L. High-Dimensional Quantile Mediation Analysis with Application to a Birth Cohort Study of Mother–Newborn Pairs. Bioinformatics. 2024. DOI: 10.1093/bioinformatics/btae055. PMID: 38290773; PMCID: PMC10873903

7. Bai X, Zheng Y, Hou L, Zheng C, Liu L, Zhang H. An Efficient Testing Procedure for High-dimensional Mediators with FDR Control. Statistics in Biosciences. 2024. DOI: 10.1007/s12561-024-09447-4.

Classic high-dimensional mediation analysis

Description

classicHIMA is used to estimate and test classic high-dimensional mediation effects (linear & logistic regression).

Usage

classicHIMA(
  X,
  M,
  Y,
  COV.XM = NULL,
  COV.MY = COV.XM,
  Y.type = c("continuous", "binary"),
  M.type = c("gaussian", "negbin"),
  penalty = c("MCP", "SCAD", "lasso"),
  topN = NULL,
  parallel = FALSE,
  ncore = 1,
  scale = TRUE,
  Bonfcut = 0.05,
  verbose = FALSE,
  ...
)
classicHIMA(
  X,
  M,
  Y,
  COV.XM = NULL,
  COV.MY = COV.XM,
  Y.type = c("continuous", "binary"),
  M.type = c("gaussian", "negbin"),
  penalty = c("MCP", "SCAD", "lasso"),
  topN = NULL,
  parallel = FALSE,
  ncore = 1,
  scale = TRUE,
  Bonfcut = 0.05,
  verbose = FALSE,
  ...
)

Arguments

`X`	a vector of exposure. Do not use data.frame or matrix.
`M`	a `data.frame` or `matrix` of high-dimensional mediators. Rows represent samples, columns represent variables.
`Y`	a vector of outcome. Can be either continuous or binary (0-1). Do not use data.frame or matrix.
`COV.XM`	a `data.frame` or `matrix` of covariates dataset for testing the association `M ~ X`. Covariates specified here will not participate penalization. Default = `NULL`. If the covariates contain mixed data types, please make sure all categorical variables are properly formatted as `factor` type.
`COV.MY`	a `data.frame` or `matrix` of covariates dataset for testing the association `Y ~ M`. Covariates specified here will not participate penalization. If not specified, the same set of covariates for `M ~ X` will be applied. Using different sets of covariates is allowed but this needs to be handled carefully.
`Y.type`	data type of outcome (`Y`). Either 'continuous' (default) or 'binary'.
`M.type`	data type of mediator (`M`). Either 'gaussian' (default) or 'negbin' (i.e., negative binomial).
`penalty`	the penalty to be applied to the model. Either 'MCP' (the default), 'SCAD', or 'lasso'.
`topN`	an integer specifying the number of top markers from sure independent screening. Default = `NULL`. If `NULL`, `topN` will be either `ceiling(n/log(n))` for continuous outcome, or `ceiling(n/(2*log(n)))` for binary outcome, where `n` is the sample size. If the sample size is greater than topN (pre-specified or calculated), all mediators will be included in the test (i.e. low-dimensional scenario).
`parallel`	logical. Enable parallel computing feature? Default = `FALSE`.
`ncore`	number of cores to run parallel computing Valid when `parallel == TRUE`. By default max number of cores available in the machine will be utilized.
`scale`	logical. Should the function scale the data? Default = `TRUE`.
`Bonfcut`	Bonferroni-corrected p value cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose? Default = `FALSE`.
`...`	other arguments passed to `ncvreg`.

Value

A data.frame containing mediation testing results of selected mediators.

Index:: mediation name of selected significant mediator.
alpha_hat:: coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
beta_hat:: coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
IDE:: mediation (indirect) effect, i.e., alpha*beta.
rimp:: relative importance of the mediator.
pmax:: joint raw p-value of selected significant mediator (based on Bonferroni method).

References

Zhang H, Zheng Y, Zhang Z, Gao T, Joyce B, Yoon G, Zhang W, Schwartz J, Just A, Colicino E, Vokonas P, Zhao L, Lv J, Baccarelli A, Hou L, Liu L. Estimating and Testing High-dimensional Mediation Effects in Epigenetic Studies. Bioinformatics. 2016. DOI: 10.1093/bioinformatics/btw351. PMID: 27357171; PMCID: PMC5048064

Examples

## Not run: 
# Note: In the following examples, M1, M2, and M3 are true mediators.
data(himaDat)

# When Y is continuous and normally distributed
# Example 1 (continuous outcome):
head(himaDat$Example1$PhenoData)

hima.fit <- classicHIMA(
  X = himaDat$Example1$PhenoData$Treatment,
  Y = himaDat$Example1$PhenoData$Outcome,
  M = himaDat$Example1$Mediator,
  COV.XM = himaDat$Example1$PhenoData[, c("Sex", "Age")],
  Y.type = "continuous",
  scale = FALSE, # Disabled only for simulation data
  verbose = TRUE
)
hima.fit

# When Y is binary
# Example 2 (binary outcome):
head(himaDat$Example2$PhenoData)

hima.logistic.fit <- classicHIMA(
  X = himaDat$Example2$PhenoData$Treatment,
  Y = himaDat$Example2$PhenoData$Disease,
  M = himaDat$Example2$Mediator,
  COV.XM = himaDat$Example2$PhenoData[, c("Sex", "Age")],
  Y.type = "binary",
  scale = FALSE, # Disabled only for simulation data
  verbose = TRUE
)
hima.logistic.fit

## End(Not run)

## Not run: 
# Note: In the following examples, M1, M2, and M3 are true mediators.
data(himaDat)

# When Y is continuous and normally distributed
# Example 1 (continuous outcome):
head(himaDat$Example1$PhenoData)

hima.fit <- classicHIMA(
  X = himaDat$Example1$PhenoData$Treatment,
  Y = himaDat$Example1$PhenoData$Outcome,
  M = himaDat$Example1$Mediator,
  COV.XM = himaDat$Example1$PhenoData[, c("Sex", "Age")],
  Y.type = "continuous",
  scale = FALSE, # Disabled only for simulation data
  verbose = TRUE
)
hima.fit

# When Y is binary
# Example 2 (binary outcome):
head(himaDat$Example2$PhenoData)

hima.logistic.fit <- classicHIMA(
  X = himaDat$Example2$PhenoData$Treatment,
  Y = himaDat$Example2$PhenoData$Disease,
  M = himaDat$Example2$Mediator,
  COV.XM = himaDat$Example2$PhenoData[, c("Sex", "Age")],
  Y.type = "binary",
  scale = FALSE, # Disabled only for simulation data
  verbose = TRUE
)
hima.logistic.fit

## End(Not run)

High-dimensional mediation analysis with de-biased lasso penalty

Description

dblassoHIMA is used to estimate and test high-dimensional mediation effects using de-biased lasso penalty.

Usage

dblassoHIMA(
  X,
  M,
  Y,
  COV = NULL,
  topN = NULL,
  scale = TRUE,
  FDRcut = 0.05,
  verbose = FALSE
)
dblassoHIMA(
  X,
  M,
  Y,
  COV = NULL,
  topN = NULL,
  scale = TRUE,
  FDRcut = 0.05,
  verbose = FALSE
)

Arguments

`X`	a vector of exposure.
`M`	a `data.frame` or `matrix` of high-dimensional mediators. Rows represent samples, columns represent variables.
`Y`	a vector of outcome. Can be either continuous or binary (0-1).
`COV`	a `data.frame` or `matrix` of covariates dataset for testing the association M ~ X and Y ~ M.
`topN`	an integer specifying the number of top markers from sure independent screening. Default = `NULL`. If `NULL`, `topN` will be `ceiling(n/log(n))`, where `n` is the sample size. If the sample size is greater than topN (pre-specified or calculated), all mediators will be included in the test (i.e. low-dimensional scenario).
`scale`	logical. Should the function scale the data? Default = `TRUE`.
`FDRcut`	HDMT pointwise FDR cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose? Default = `FALSE`.

Value

A data.frame containing mediation testing results of significant mediators (FDR <FDRcut).

Index:: mediation name of selected significant mediator.
alpha_hat:: coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
alpha_se:: standard error for alpha.
beta_hat:: coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
beta_se:: standard error for beta.
IDE:: mediation (indirect) effect, i.e., alpha*beta.
rimp:: relative importance of the mediator.
pmax:: joint raw p-value of selected significant mediator (based on HDMT pointwise FDR method).

References

Perera C, Zhang H, Zheng Y, Hou L, Qu A, Zheng C, Xie K, Liu L. HIMA2: high-dimensional mediation analysis and its application in epigenome-wide DNA methylation data. BMC Bioinformatics. 2022. DOI: 10.1186/s12859-022-04748-1. PMID: 35879655; PMCID: PMC9310002

Examples

## Not run: 
# Note: In the following examples, M1, M2, and M3 are true mediators.
data(himaDat)

# Y is continuous and normally distributed
# Example:
head(himaDat$Example1$PhenoData)

dblassohima.fit <- dblassoHIMA(
  X = himaDat$Example1$PhenoData$Treatment,
  Y = himaDat$Example1$PhenoData$Outcome,
  M = himaDat$Example1$Mediator,
  COV = himaDat$Example1$PhenoData[, c("Sex", "Age")],
  scale = FALSE, # Disabled only for simulation data
  FDRcut = 0.05,
  verbose = TRUE
)
dblassohima.fit

## End(Not run)

## Not run: 
# Note: In the following examples, M1, M2, and M3 are true mediators.
data(himaDat)

# Y is continuous and normally distributed
# Example:
head(himaDat$Example1$PhenoData)

dblassohima.fit <- dblassoHIMA(
  X = himaDat$Example1$PhenoData$Treatment,
  Y = himaDat$Example1$PhenoData$Outcome,
  M = himaDat$Example1$Mediator,
  COV = himaDat$Example1$PhenoData[, c("Sex", "Age")],
  scale = FALSE, # Disabled only for simulation data
  FDRcut = 0.05,
  verbose = TRUE
)
dblassohima.fit

## End(Not run)

Efficient high-dimensional mediation analysis

Description

eHIMA is used to estimate and test high-dimensional mediation effects using an efficient algorithm. It provides higher statistical power than the standard hima. Note: efficient HIMA is only applicable to mediators and outcomes that are both continuous and normally distributed.

Usage

eHIMA(
  X,
  M,
  Y,
  COV = NULL,
  topN = NULL,
  scale = TRUE,
  FDRcut = 0.05,
  verbose = FALSE
)
eHIMA(
  X,
  M,
  Y,
  COV = NULL,
  topN = NULL,
  scale = TRUE,
  FDRcut = 0.05,
  verbose = FALSE
)

Arguments

`X`	a vector of exposure.
`M`	a `data.frame` or `matrix` of high-dimensional mediators. Rows represent samples, columns represent mediator variables. `M` has to be continuous and normally distributed.
`Y`	a vector of continuous outcome. Do not use data.frame or matrix.
`COV`	a matrix of adjusting covariates. Rows represent samples, columns represent variables. Can be `NULL`.
`topN`	an integer specifying the number of top markers from sure independent screening. Default = `NULL`. If `NULL`, `topN` will be `2*ceiling(n/log(n))`, where `n` is the sample size. If the sample size is greater than topN (pre-specified or calculated), all mediators will be included in the test (i.e. low-dimensional scenario).
`scale`	logical. Should the function scale the data? Default = `TRUE`.
`FDRcut`	Benjamini-Hochberg FDR cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose? Default = `FALSE`.

Value

A data.frame containing mediation testing results of significant mediators (FDR <FDRcut).

Index:: mediation name of selected significant mediator.
alpha_hat:: coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
alpha_se:: standard error for alpha.
beta_hat:: coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
beta_se:: standard error for beta.
IDE:: mediation (indirect) effect, i.e., alpha*beta.
rimp:: relative importance of the mediator.
pmax:: joint raw p-value of selected significant mediator (based on divide-aggregate composite-null test [DACT] method).

References

Bai X, Zheng Y, Hou L, Zheng C, Liu L, Zhang H. An Efficient Testing Procedure for High-dimensional Mediators with FDR Control. Statistics in Biosciences. 2024. DOI: 10.1007/s12561-024-09447-4.

Examples

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

# Y is continuous and normally distributed
# Example (continuous outcome):
head(himaDat$Example1$PhenoData)

eHIMA.fit <- eHIMA(
  X = himaDat$Example1$PhenoData$Treatment,
  Y = himaDat$Example1$PhenoData$Outcome,
  M = himaDat$Example1$Mediator,
  COV = himaDat$Example1$PhenoData[, c("Sex", "Age")],
  scale = FALSE, # Disabled only for simulation data
  FDRcut = 0.05,
  verbose = TRUE
)
eHIMA.fit

## End(Not run)

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

# Y is continuous and normally distributed
# Example (continuous outcome):
head(himaDat$Example1$PhenoData)

eHIMA.fit <- eHIMA(
  X = himaDat$Example1$PhenoData$Treatment,
  Y = himaDat$Example1$PhenoData$Outcome,
  M = himaDat$Example1$Mediator,
  COV = himaDat$Example1$PhenoData[, c("Sex", "Age")],
  scale = FALSE, # Disabled only for simulation data
  FDRcut = 0.05,
  verbose = TRUE
)
eHIMA.fit

## End(Not run)

Example datasets for HIMA demo

Description

A list containing datasets for various scenarios of HIMA. Each dataset contains a phenotype data frame and a high-dimension mediator data matrix. The datasets are simulated using parameters generated from real datasets. The code used to generate the data can be found in "/inst/script" folder of the source package.

Usage

himaDat
himaDat

Format

An object of class list of length 5.

Value

A list of example datasets for HIMA demo and testing.

–>> Example dataset 1 for HIMA: Continuous outcome <<–

Treatment:: treated (value = 1) or not treated (value = 0)
Outcome:: outcome of the treatment- a normally distributed continuous variable
Sex:: female (value = 1) or male (value = 0)
Age:: Age of the participant

–>> Example dataset 2 for HIMA: Binary outcome <<–

Treatment:: treated (value = 1) or not treated (value = 0)
Disease:: diseased (value = 1) or healthy (value = 0)
Sex:: female (value = 1) or male (value = 0)
Age:: Age of the participant

–>> Example dataset 3 for HIMA: Survival data outcome <<–

Treatment:: treated (value = 1) or not treated (value = 0)
Status:: Status indicator: dead (value = 1) or alive (value = 0)
Time:: time to event
Sex:: female (value = 1) or male (value = 0)
Age:: Age of the participant

–>> Example dataset 4 for HIMA: Compositional mediator (e.g., microbiome) <<–

Treatment:: treated (value = 1) or not treated (value = 0)
Outcome:: outcome of the treatment- a normally distributed continuous variable
Sex:: female (value = 1) or male (value = 0)
Age:: Age of the participant

–>> Example dataset 5 for HIMA: High-dimensional quantile mediation analysis <<–

Treatment:: treated (value = 1) or not treated (value = 0)
Outcome:: outcome of the treatment- abnormally distributed continuous variable
Sex:: female (value = 1) or male (value = 0)
Age:: Age of the participant

High-dimensional Mediation Analysis

Description

himaFit is a wrapper function designed to perform various HIMA methods for estimating and testing high-dimensional mediation effects. himaFit can automatically select the appropriate HIMA method based on the outcome and mediator data type.

Usage

himaFit(
  formula,
  data.pheno,
  data.M,
  mediator.type = c("gaussian", "negbin", "compositional"),
  penalty = c("DBlasso", "MCP", "SCAD", "lasso"),
  quantile = FALSE,
  efficient = FALSE,
  scale = TRUE,
  Sigcut = 0.05,
  verbose = FALSE,
  ...
)
himaFit(
  formula,
  data.pheno,
  data.M,
  mediator.type = c("gaussian", "negbin", "compositional"),
  penalty = c("DBlasso", "MCP", "SCAD", "lasso"),
  quantile = FALSE,
  efficient = FALSE,
  scale = TRUE,
  Sigcut = 0.05,
  verbose = FALSE,
  ...
)

Arguments

`formula`	an object of class `formula`: a symbolic description of the overall effect model, i.e., `outcome ~ exposure + covariates`, to be fitted. Make sure the "exposure" is the variable of interest, which must be listed as the first variable in the right hand side of the formula.
`data.pheno`	a data frame containing exposure and covariates that are listed in the right hand side of the `formula`. The variable names must match those listed in `formula`. By default `himaFit` will scale `data.pheno`.
`data.M`	a `data.frame` or `matrix` of high-dimensional mediators. Rows represent samples, columns represent variables. By default `himaFit` will scale `data.M`.
`mediator.type`	data type of high-dimensional mediators (`data.M`). Either `'gaussian'` (default, for continuous mediators), `'negbin'` (i.e., negative binomial, for RNA-seq data as mediators), or `'compositional'` (for microbiome data as mediators).
`penalty`	the penalty to be applied to the model. Either `'DBlasso'` (De-biased LASSO, default), `'MCP'`, `'SCAD'`, or `'lasso'`. Please note, survival HIMA and microbiome HIMA can be only performed with `'DBlasso'`; Quantile HIMA cannot be performed with `'DBlasso'`. Not applicable for efficient HIMA (when `efficient = TRUE`), as it will always apply `'MCP'`.
`quantile`	use quantile HIMA (`qHIMA`). Only applicable for classic HIMA with continues outcome. Please add parameter `tau` to specify the desired quantile (can be a vector), otherwise the default `tau = 0.5` will be used.
`efficient`	use efficient HIMA (`eHIMA`). Only applicable for classic HIMA with continuous outcome. and `mediator.type = "gaussian"`). Default = `FALSE`.
`scale`	logical. Should the function scale the data (exposure, mediators, and covariates)? Default = `TRUE`. Note: for simulation study, scale can be turned off to avoid estimate compression.
`Sigcut`	cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose and show the progression? Default = `FALSE`.
`...`	reserved passing parameter.

Value

A data.frame containing mediation testing results of selected mediators.

ID:: Mediator ID/name.
alpha:: Coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
beta:: Coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
alpha*beta:: Mediation (indirect) effect.
Relative Importance:: Relative importance of the mediator. It is the proportion of the mediation effect for each mediator out of the sum of the mediation effect (absolute value) across all significant mediators selected.
p-value:: Joint raw p-value of significant mediators selected based on corresponding approach.
tau:: Quantile level of the outcome (applicable only to the quantile mediation model).

References

4. Zhang H, Chen J, Li Z, Liu L. Testing for Mediation Effect with Application to Human Microbiome Data. Stat Biosci. 2021. DOI: 10.1007/s12561-019-09253-3. PMID: 34093887; PMCID: PMC8177450

7. Bai X, Zheng Y, Hou L, Zheng C, Liu L, Zhang H. An Efficient Testing Procedure for High-dimensional Mediators with FDR Control. Statistics in Biosciences. 2024. DOI: 10.1007/s12561-024-09447-4.

Examples

## Not run: 
# Note: In the following examples, M1, M2, and M3 are true mediators.
data(himaDat)

# Example 1 (continuous outcome - linear HIMA):
head(himaDat$Example1$PhenoData)

e1 <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example1$PhenoData,
  data.M = himaDat$Example1$Mediator,
  mediator.type = "gaussian",
  penalty = "MCP", # Can be "DBlasso" for dblassoHIMA
  scale = FALSE
) # Disabled only for simulation data
e1
attributes(e1)$variable.labels

# Efficient HIMA (only applicable to mediators and outcomes that are
# both continuous and normally distributed.)
e1e <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example1$PhenoData,
  data.M = himaDat$Example1$Mediator,
  mediator.type = "gaussian",
  efficient = TRUE,
  scale = FALSE
) # Disabled only for simulation data
e1e
attributes(e1e)$variable.labels

# Example 2 (binary outcome - logistic HIMA):
head(himaDat$Example2$PhenoData)

e2 <- himaFit(Disease ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example2$PhenoData,
  data.M = himaDat$Example2$Mediator,
  mediator.type = "gaussian",
  penalty = "MCP",
  scale = FALSE
) # Disabled only for simulation data
e2
attributes(e2)$variable.labels

# Example 3 (time-to-event outcome - survival HIMA):
head(himaDat$Example3$PhenoData)

e3 <- himaFit(Surv(Status, Time) ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example3$PhenoData,
  data.M = himaDat$Example3$Mediator,
  mediator.type = "gaussian",
  penalty = "DBlasso",
  scale = FALSE
) # Disabled only for simulation data
e3
attributes(e3)$variable.labels

# Example 4 (compositional data as mediator, e.g., microbiome):
head(himaDat$Example4$PhenoData)

e4 <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example4$PhenoData,
  data.M = himaDat$Example4$Mediator,
  mediator.type = "compositional",
  penalty = "DBlasso"
) # Scaling is always enabled internally for microHIMA
e4
attributes(e4)$variable.labels

#' # Example 5 (quantile mediation anlaysis - quantile HIMA):
head(himaDat$Example5$PhenoData)

# Note that the function will prompt input for quantile level.
e5 <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example5$PhenoData,
  data.M = himaDat$Example5$Mediator,
  mediator.type = "gaussian",
  quantile = TRUE,
  penalty = "MCP", # Quantile HIMA does not support DBlasso
  scale = FALSE, # Disabled only for simulation data
  tau = c(0.3, 0.5, 0.7)
) # Specify multiple quantile level
e5
attributes(e5)$variable.labels

## End(Not run)

## Not run: 
# Note: In the following examples, M1, M2, and M3 are true mediators.
data(himaDat)

# Example 1 (continuous outcome - linear HIMA):
head(himaDat$Example1$PhenoData)

e1 <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example1$PhenoData,
  data.M = himaDat$Example1$Mediator,
  mediator.type = "gaussian",
  penalty = "MCP", # Can be "DBlasso" for dblassoHIMA
  scale = FALSE
) # Disabled only for simulation data
e1
attributes(e1)$variable.labels

# Efficient HIMA (only applicable to mediators and outcomes that are
# both continuous and normally distributed.)
e1e <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example1$PhenoData,
  data.M = himaDat$Example1$Mediator,
  mediator.type = "gaussian",
  efficient = TRUE,
  scale = FALSE
) # Disabled only for simulation data
e1e
attributes(e1e)$variable.labels

# Example 2 (binary outcome - logistic HIMA):
head(himaDat$Example2$PhenoData)

e2 <- himaFit(Disease ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example2$PhenoData,
  data.M = himaDat$Example2$Mediator,
  mediator.type = "gaussian",
  penalty = "MCP",
  scale = FALSE
) # Disabled only for simulation data
e2
attributes(e2)$variable.labels

# Example 3 (time-to-event outcome - survival HIMA):
head(himaDat$Example3$PhenoData)

e3 <- himaFit(Surv(Status, Time) ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example3$PhenoData,
  data.M = himaDat$Example3$Mediator,
  mediator.type = "gaussian",
  penalty = "DBlasso",
  scale = FALSE
) # Disabled only for simulation data
e3
attributes(e3)$variable.labels

# Example 4 (compositional data as mediator, e.g., microbiome):
head(himaDat$Example4$PhenoData)

e4 <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example4$PhenoData,
  data.M = himaDat$Example4$Mediator,
  mediator.type = "compositional",
  penalty = "DBlasso"
) # Scaling is always enabled internally for microHIMA
e4
attributes(e4)$variable.labels

#' # Example 5 (quantile mediation anlaysis - quantile HIMA):
head(himaDat$Example5$PhenoData)

# Note that the function will prompt input for quantile level.
e5 <- himaFit(Outcome ~ Treatment + Sex + Age,
  data.pheno = himaDat$Example5$PhenoData,
  data.M = himaDat$Example5$Mediator,
  mediator.type = "gaussian",
  quantile = TRUE,
  penalty = "MCP", # Quantile HIMA does not support DBlasso
  scale = FALSE, # Disabled only for simulation data
  tau = c(0.3, 0.5, 0.7)
) # Specify multiple quantile level
e5
attributes(e5)$variable.labels

## End(Not run)

High-dimensional mediation analysis for compositional microbiome data

Description

microHIMA is used to estimate and test high-dimensional mediation effects for compositional microbiome data.

Usage

microHIMA(X, OTU, Y, COV = NULL, FDRcut = 0.05, verbose = FALSE)
microHIMA(X, OTU, Y, COV = NULL, FDRcut = 0.05, verbose = FALSE)

Arguments

`X`	a vector of exposure.
`OTU`	a `data.frame` or `matrix` of high-dimensional Operational Taxonomic Unit (OTU) data (mediators). Rows represent samples, columns represent variables.
`Y`	a vector of continuous outcome. Binary outcome is not allowed.
`COV`	a `data.frame` or `matrix` of adjusting covariates. Rows represent samples, columns represent microbiome variables. Can be `NULL`.
`FDRcut`	Hommel FDR cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose? Default = `FALSE`.

Value

A data.frame containing mediation testing results of significant mediators (FDR <FDRcut).

Index:: mediation name of selected significant mediator.
alpha_hat:: coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
alpha_se:: standard error for alpha.
beta_hat:: coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
beta_se:: standard error for beta.
IDE:: mediation (indirect) effect, i.e., alpha*beta.
rimp:: relative importance of the mediator.
pmax:: joint raw p-value of selected significant mediator (based on Hommel FDR method).

References

1. Zhang H, Chen J, Feng Y, Wang C, Li H, Liu L. Mediation effect selection in high-dimensional and compositional microbiome data. Stat Med. 2021. DOI: 10.1002/sim.8808. PMID: 33205470; PMCID: PMC7855955

2. Zhang H, Chen J, Li Z, Liu L. Testing for mediation effect with application to human microbiome data. Stat Biosci. 2021. DOI: 10.1007/s12561-019-09253-3. PMID: 34093887; PMCID: PMC8177450

Examples

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

head(himaDat$Example4$PhenoData)

microHIMA.fit <- microHIMA(
  X = himaDat$Example4$PhenoData$Treatment,
  Y = himaDat$Example4$PhenoData$Outcome,
  OTU = himaDat$Example4$Mediator,
  COV = himaDat$Example4$PhenoData[, c("Sex", "Age")],
  FDRcut = 0.05,
  verbose = TRUE
)
microHIMA.fit

## End(Not run)

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

head(himaDat$Example4$PhenoData)

microHIMA.fit <- microHIMA(
  X = himaDat$Example4$PhenoData$Treatment,
  Y = himaDat$Example4$PhenoData$Outcome,
  OTU = himaDat$Example4$Mediator,
  COV = himaDat$Example4$PhenoData[, c("Sex", "Age")],
  FDRcut = 0.05,
  verbose = TRUE
)
microHIMA.fit

## End(Not run)

High-dimensional quantile mediation analysis

Description

qHIMA is used to estimate and test high-dimensional quantile mediation effects.

Usage

qHIMA(
  X,
  M,
  Y,
  COV = NULL,
  penalty = c("MCP", "SCAD", "lasso"),
  topN = NULL,
  tau = 0.5,
  scale = TRUE,
  Bonfcut = 0.05,
  verbose = FALSE,
  ...
)
qHIMA(
  X,
  M,
  Y,
  COV = NULL,
  penalty = c("MCP", "SCAD", "lasso"),
  topN = NULL,
  tau = 0.5,
  scale = TRUE,
  Bonfcut = 0.05,
  verbose = FALSE,
  ...
)

Arguments

`X`	a vector of exposure.
`M`	a `data.frame` or `matrix` of high-dimensional mediators. Rows represent samples, columns represent mediator variables.
`Y`	a vector of continuous outcome. Do not use data.frame or matrix.
`COV`	a matrix of adjusting covariates. Rows represent samples, columns represent variables. Can be `NULL`.
`penalty`	the penalty to be applied to the model (a parameter passed to function `conquer.cv.reg` in package `conquer`. Either 'MCP' (the default), 'SCAD', or 'lasso'.
`topN`	an integer specifying the number of top markers from sure independent screening. Default = `NULL`. If `NULL`, `topN` will be `2*ceiling(n/log(n))`, where `n` is the sample size. If the sample size is greater than topN (pre-specified or calculated), all mediators will be included in the test (i.e. low-dimensional scenario).
`tau`	quantile level of outcome. Default = 0.5. A vector of tau is accepted.
`scale`	logical. Should the function scale the data? Default = `TRUE`.
`Bonfcut`	Bonferroni-corrected p value cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose? Default = `FALSE`.
`...`	reserved passing parameter.

Value

A data.frame containing mediation testing results of selected mediators (Bonferroni-adjusted p value <Bonfcut).

Index:: mediation name of selected significant mediator.
alpha_hat:: coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
alpha_se:: standard error for alpha.
beta_hat:: coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
beta_se:: standard error for beta.
IDE:: mediation (indirect) effect, i.e., alpha*beta.
rimp:: relative importance of the mediator.
pmax:: joint raw p-value of selected significant mediator (based on Bonferroni method).

References

Zhang H, Hong X, Zheng Y, Hou L, Zheng C, Wang X, Liu L. High-Dimensional Quantile Mediation Analysis with Application to a Birth Cohort Study of Mother–Newborn Pairs. Bioinformatics. 2024. DOI: 10.1093/bioinformatics/btae055. PMID: 38290773; PMCID: PMC10873903

Examples

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

head(himaDat$Example5$PhenoData)

qHIMA.fit <- qHIMA(
  X = himaDat$Example5$PhenoData$Treatment,
  M = himaDat$Example5$Mediator,
  Y = himaDat$Example5$PhenoData$Outcome,
  COV = himaDat$Example5$PhenoData[, c("Sex", "Age")],
  tau = c(0.3, 0.5, 0.7),
  scale = FALSE, # Disabled only for simulation data
  Bonfcut = 0.05,
  verbose = TRUE
)
qHIMA.fit

## End(Not run)

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

head(himaDat$Example5$PhenoData)

qHIMA.fit <- qHIMA(
  X = himaDat$Example5$PhenoData$Treatment,
  M = himaDat$Example5$Mediator,
  Y = himaDat$Example5$PhenoData$Outcome,
  COV = himaDat$Example5$PhenoData[, c("Sex", "Age")],
  tau = c(0.3, 0.5, 0.7),
  scale = FALSE, # Disabled only for simulation data
  Bonfcut = 0.05,
  verbose = TRUE
)
qHIMA.fit

## End(Not run)

High-dimensional mediation analysis for survival data

Description

survHIMA is used to estimate and test high-dimensional mediation effects for survival data.

Usage

survHIMA(
  X,
  M,
  OT,
  status,
  COV = NULL,
  topN = NULL,
  scale = TRUE,
  FDRcut = 0.05,
  verbose = FALSE
)
survHIMA(
  X,
  M,
  OT,
  status,
  COV = NULL,
  topN = NULL,
  scale = TRUE,
  FDRcut = 0.05,
  verbose = FALSE
)

Arguments

`X`	a vector of exposure.
`M`	a `data.frame` or `matrix` of high-dimensional mediators. Rows represent samples, columns represent mediator variables.
`OT`	a vector of observed failure times.
`status`	a vector of censoring indicator (`status = 1`: uncensored; `status = 0`: censored)
`COV`	a matrix of adjusting covariates. Rows represent samples, columns represent variables. Can be `NULL`.
`topN`	an integer specifying the number of top markers from sure independent screening. Default = `NULL`. If `NULL`, `topN` will be `ceiling(n/log(n))`, where `n` is the sample size. If the sample size is greater than topN (pre-specified or calculated), all mediators will be included in the test (i.e. low-dimensional scenario).
`scale`	logical. Should the function scale the data? Default = `TRUE`.
`FDRcut`	HDMT pointwise FDR cutoff applied to select significant mediators. Default = `0.05`.
`verbose`	logical. Should the function be verbose? Default = `FALSE`.

Value

A data.frame containing mediation testing results of significant mediators (FDR <FDRcut).

Index:: mediation name of selected significant mediator.
alpha_hat:: coefficient estimates of exposure (X) –> mediators (M) (adjusted for covariates).
alpha_se:: standard error for alpha.
beta_hat:: coefficient estimates of mediators (M) –> outcome (Y) (adjusted for covariates and exposure).
beta_se:: standard error for beta.
IDE:: mediation (indirect) effect, i.e., alpha*beta.
rimp:: relative importance of the mediator.
pmax:: joint raw p-value of selected significant mediator (based on HDMT pointwise FDR method).

References

Zhang H, Zheng Y, Hou L, Zheng C, Liu L. Mediation Analysis for Survival Data with High-Dimensional Mediators. Bioinformatics. 2021. DOI: 10.1093/bioinformatics/btab564. PMID: 34343267; PMCID: PMC8570823

Examples

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

head(himaDat$Example3$PhenoData)

survHIMA.fit <- survHIMA(
  X = himaDat$Example3$PhenoData$Treatment,
  M = himaDat$Example3$Mediator,
  OT = himaDat$Example3$PhenoData$Time,
  status = himaDat$Example3$PhenoData$Status,
  COV = himaDat$Example3$PhenoData[, c("Sex", "Age")],
  scale = FALSE, # Disabled only for simulation data
  FDRcut = 0.05,
  verbose = TRUE
)
survHIMA.fit

## End(Not run)

## Not run: 
# Note: In the following example, M1, M2, and M3 are true mediators.
data(himaDat)

head(himaDat$Example3$PhenoData)

survHIMA.fit <- survHIMA(
  X = himaDat$Example3$PhenoData$Treatment,
  M = himaDat$Example3$Mediator,
  OT = himaDat$Example3$PhenoData$Time,
  status = himaDat$Example3$PhenoData$Status,
  COV = himaDat$Example3$PhenoData[, c("Sex", "Age")],
  scale = FALSE, # Disabled only for simulation data
  FDRcut = 0.05,
  verbose = TRUE
)
survHIMA.fit

## End(Not run)

Package 'HIMA'

Help Index

High-Dimensional Mediation Analysis for 'Omic' Data

Description

Details

Author(s)

References

Classic high-dimensional mediation analysis

Description

Usage

Arguments

Value

References

Examples

High-dimensional mediation analysis with de-biased lasso penalty

Description

Usage

Arguments

Value

References

Examples

Efficient high-dimensional mediation analysis

Description

Usage

Arguments

Value

References

Examples

Example datasets for HIMA demo

Description

Usage

Format

Value

High-dimensional Mediation Analysis

Description

Usage

Arguments

Value

References

Examples

High-dimensional mediation analysis for compositional microbiome data

Description

Usage

Arguments

Value

References

Examples

High-dimensional quantile mediation analysis

Description

Usage

Arguments

Value

References

Examples

High-dimensional mediation analysis for survival data

Description

Usage

Arguments

Value

References

Examples