PathModel: Test how bacterial taxa and host variables together influence...

In katrikorpela/mare: Microbiota Analysis in R Easily

Test how bacterial taxa and host variables together influence the response variable

Description

Combine information on bacterial taxa and a set of host variables into one model for the response variable. PathModel finds the bacterial taxa and host variables that together are significantly associated with the response variable, and the host variables that are associated with the response-variable-associated bacterial taxa.

Usage

PathModel(taxonomic.table, meta, response = NULL, 
                      binomial.response = F, count.response = F, variables, readcount.cutoff = 0, 
                      outlier.cutoff = 3, 
                      select.by = NULL, select = NULL, pdf = F, 
                      min.prevalence = 0, min.abundance = 0, keep.result = F, k.value = 3.5,
                      logtrans = F, relative = T)

Arguments

taxonomic.table	Taxonomic table containing the bacterial taxa to be tested. Should be the organised file returned by function Organise.
meta	Metadata file.
response	The response variable. This has to be a numerical variable: binary (0s and 1s), counts of observations, or continuous. If the response variable is omitted, the model will use all bacterial taxa as the response variables, and finds the best total model for each taxon.
binomial.response	Set to TRUE if the response variable contains only 0s and 1s.
count.response	Set to TRUE if the response variable is measured as counts of observations.
variables	Indices or names of the metadata file that you want to use as additional explanatory variables in the model. These must be numerical, categorical variables cannot be used.
readcount.cutoff	Lowest acceptable read count per sample. Samples with fewer reads are ignored.
outlier.cutoff	Highest acceptable standard deviation from the overall mean abundance for a given taxon. Data points exceeding this value will be replaced by the cut-off value to eliminate disproportional effect of outliers on the results.
min.prevalence	Minimum acceptable prevalence of a taxon. Taxa occurring less frequently will be ignored.
min.abundance	Minimum acceptable relative abundance of a taxon. Taxa occurring at a lower relative abundance will be considered absent when calculating prevalence.
select.by	Name of a variable in the metadata file by which a subset will be selected for plotting.
select	Determines which value on the selection variable will be selected.
pdf	Should be the result figure be printed as PDF? If yes, set to TRUE.
k.value	Choose the k-value (penalty parameter) for AIC-based model reduction. The higher the value, the fewer variables will be left in the model.
logtrans	Log-transform the microbiota data?
relative	The data consist of relative abundances, i.e. have not been transformed into absolute abundances. This means the number of reads per sample will be used as an offset in the models. TRUE or FALSE.
keep.result	Store the results as a list. The first item on the list is the response model, and the second item is a list of all the bacterial models. The resulting final best models can then be accessed and inspected by using the summary command.

Details

PathModel first fits a linear model (make sure your variable is normally distributed) or a binomial model (for 0/1 response variables) or a negative binomial model (for count variables) separately to all bacterial taxa and host variables, then selects the best of those into a combined model and performs model reduction using AIC to simplify the model. It then performs the same model selection for each of the bacterial taxa retained in response variable model, now fitting negative binomial models. Then the best overall model is plotted, including the bacterial taxa and host variables associated with the response variable, and the host variables associated with these bacterial taxa.

Author(s)

Katri Korpela

katrikorpela/mare documentation built on July 17, 2022, 2:49 a.m.