filterProfileData: Filter phylogentic profiles

In BIONF/PhyloProfile: PhyloProfile

Filter phylogentic profiles

Description

Create a filtered data needed for plotting or clustering phylogenetic profiles. NOTE: this function require some intermediate steps using the results from other functions. If you would like to get a full processed data from the raw input, please use the function fromInputToProfile() instead!

Usage

filterProfileData(DF, taxaCount, refTaxon = NULL,
    percentCO = c(0, 1), coorthoCOMax = 9999,
    var1CO  = c(0, 1), var2CO = c(0, 1), var1Rel = "protein",
    var2Rel = "protein", groupByCat = FALSE, catDt = NULL,
    var1AggregateBy = "max", var2AggregateBy = "max")

Arguments

DF	a reduced dataframe contains info for all phylogenetic profiles in the selected taxonomy rank.
taxaCount	dataframe counting present taxa in each supertaxon
refTaxon	selected reference taxon. NOTE: This taxon will not be affected by the filtering. If you want to filter all, set refTaxon <- NULL. Default = NULL.
percentCO	min and max cutoffs for percentage of species present in a supertaxon. Default = c(0, 1).
coorthoCOMax	maximum number of co-orthologs allowed. Default = 9999.
var1CO	min and max cutoffs for var1. Default = c(0, 1).
var2CO	min anc max cutoffs for var2. Default = c(0, 1).
var1Rel	relation of var1 ("protein" for protein-protein or "species" for protein-species). Default = "protein".
var2Rel	relation of var2 ("protein" for protein-protein or "species" for protein-species). Default = "protein".
groupByCat	group genes by their categories (TRUE or FALSE). Default = FALSE.
catDt	dataframe contains gene categories (optional, NULL if groupByCat = FALSE or no info provided). Default = NULL.
var1AggregateBy	aggregate method for VAR1 (max, min, mean or median), applied for calculating var1 of supertaxa. Default = "max".
var2AggregateBy	aggregate method for VAR2 (max, min, mean or median), applied for calculating var2 of supertaxa. Default = "max".

Value

A filtered dataframe for generating profile plot including seed gene IDs (or orthologous group IDs), their ortholog IDs and the corresponding (super)taxa, (super)taxon IDs, number of co-orthologs in each (super)taxon, values for two additional variables var1, var2, supertaxon, and the categories of seed genes (or ortholog groups).

Author(s)

Vinh Tran [e-mail]

See Also

parseInfoProfile and reduceProfile for generating input dataframe, fullProcessedProfile for a demo full processed profile dataframe, fromInputToProfile for generating fully processed data from raw input.

Examples

# NOTE: this function require some intermediate steps using the results from
# other functions. If you would like to get a full processed data from the
# raw input, please use the function fromInputToProfile() instead!
library(dplyr)
data("fullProcessedProfile", package="PhyloProfile")
rankName <- "class"
refTaxon <- "Mammalia"
percentCutoff <- c(0.0, 1.0)
coorthologCutoffMax <- 10
var1Cutoff <- c(0.75, 1.0)
var2Cutoff <- c(0.5, 1.0)
var1Relation <- "protein"
var2Relation <- "species"
groupByCat <- FALSE
catDt <- NULL
var1AggregateBy <- "max"
var2AggregateBy <- "max"
taxonIDs <- levels(as.factor(fullProcessedProfile$ncbiID))
sortedInputTaxa <- sortInputTaxa(
    taxonIDs, rankName, refTaxon, NULL, NULL
)
taxaCount <- sortedInputTaxa %>% dplyr::group_by(supertaxon) %>%
    summarise(n = n(), .groups = "drop")
filterProfileData(
    fullProcessedProfile,
    taxaCount,
    refTaxon,
    percentCutoff,
    coorthologCutoffMax,
    var1Cutoff,
    var2Cutoff,
    var1Relation,
    var2Relation,
    groupByCat,
    catDt,
    var1AggregateBy,
    var2AggregateBy
)

BIONF/PhyloProfile documentation built on Dec. 18, 2024, 7:33 a.m.