poison_regression_permute: poison_regression_permute

In dlemas/microbes: tools for analyzing microbiome data

Description

[TEXT]

Usage

poison_regression_permute(wgs_path_count, n.total, n.gr1, n.gr2, email)

Arguments

wgs_path_count	[TEXT]
n.total	[TEXT]
n.gr1	[TEXT]
n.gr2	[TEXT]
email	[TEXT]

Details

[TEXT]

Value

[TEXT]

Note

no further notes

Author(s)

dlemas

References

[TEXT]

See Also

[TEXT]

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (wgs_path_count, n.total, n.gr1, n.gr2, email) 
{
    library(gtools)
    library(mail)
    ptm <- proc.time()
    df.pathway.new = wgs_path_count[, -1]
    rownames(df.pathway.new) = NULL
    colnames(df.pathway.new) = NULL
    data2 = data.matrix(df.pathway.new)
    Y <- c(rep(1, n.gr1), rep(0, n.gr2))
    I <- combinations(n = n.total, r = n.gr2, set = FALSE, repeats.allowed = FALSE)
    permY <- matrix(nrow = nrow(I), ncol = n.total, 0)
    for (i in 1:nrow(permY)) permY[i, I[i, ]] <- 1
    PVALUES <- NULL
    for (i in 1:nrow(data2)) {
        ans <- summary(glm(data2[i, ] ~ Y, family = poisson(link = "log")))
        obs.LRTS <- ans$null.deviance - ans$deviance
        perm.LRTS <- apply(permY, MARGIN = 1, function(x) {
            ans <- summary(glm(data2[i, ] ~ x, family = poisson(link = "log")))
            ans$null.deviance - ans$deviance
        })
        PVALUES <- c(PVALUES, mean(perm.LRTS >= obs.LRTS))
    }
    stop = c(proc.time() - ptm)
    stop[1]
    message.detail = paste("Calculation finished. Total Run Time: ", 
        round(stop[1], digits = 2), " seconds", sep = "")
    email.notification = sendmail(email, subject = "Notification from R: poison_regression_permute complete", 
        message = message.detail, password = "rmail")
    return(wgs_path_count.new = cbind(wgs_path_count, PVALUES))
    return(email.notification)
  }

dlemas/microbes documentation built on May 15, 2019, 9:15 a.m.