# Calculate Attenuation from AUC and Other Transcript Features This function computes the attenuation values for each window of each transcript based on the data frames obtained with the functions 'allauc', 'kneeid', and 'countna'. ```r attenuation(allaucdf, kneedf, matnatrans, bytranslistmean, expdf, dfmeandiff, nbcpu = 1, significant = FALSE, replaceval = NA, pval = 0.1, showtime = FALSE, verbose = TRUE) ``` ## Arguments - `allaucdf`: A data frame containing AUC results for transcripts (see allauc). - `kneedf`: A data frame containing the inflection points (see kneeid). - `matnatrans`: A data frame containing the number of missing values per transcript (see countna). - `bytranslistmean`: A list of data frames with mean values by transcripts. - `expdf`: A data frame containing experiment data that should have columns named 'condition', 'replicate', 'strand', and 'path'. - `dfmeandiff`: A data frame containing means and differences in mean values, if more than one condition. (see meandifference). - `nbcpu`: An integer specifying the number of CPU cores to use for parallel processing. The parallelization is done on bytranslistmean whose number of elements is equal to the number of lines provided as input of 'averageandfilterexprs'. Defaults to `1`. - `significant`: A logical indicating whether to filter out non-significant attenuation values. Defaults to `FALSE`. - `replaceval`: A value to replace non-significant attenuation values Defaults to `NA`. - `pval`: A numeric value specifying the p-value threshold for significance of the KS test. Defaults to `0.1`. - `showtime`: A logical value indicating if the duration of the function processing should be indicated before ending. Defaults to `FALSE`. - `verbose`: A logical value indicating whether to print progress messages Defaults to `TRUE`. ## Returns A data frame containing the computed attenuation values along with associated transcript information. ## Details The function merges several data frames to create a comprehensive dataset for each transcript. It computes mean values for the "up" and "down" segments of the transcript. The direction is determined by comparing the coordinates to the knee values. up = coord < knee and down = coord > knee. The up and down indexes are then retrieved and the attenuation scores are computed as: att <- 100 - downmean / upmean * 100 ## Examples ```r exppath <- system.file("extdata", "exptab.csv", package="tepr") transpath <- system.file("extdata", "cugusi_6.tsv", package="tepr") expthres <- 0.1 ## Reading tables expdf <- read.csv(exppath) transdf <- read.delim(transpath, header = FALSE) ## Computing intermediate steps avfilt <- averageandfilterexprs(expdf, transdf, expthres, showtime = FALSE, verbose = FALSE) rescountna <- countna(avfilt, expdf, nbcpu = 1, verbose = FALSE) ecdf <- genesECDF(avfilt, verbose = FALSE) resecdf <- ecdf[[1]] nbwindows <- ecdf[[2]] resmeandiff <- meandifference(resecdf, expdf, nbwindows, verbose = FALSE) bytranslistmean <- split(resmeandiff, factor(resmeandiff$transcript)) resknee <- kneeid(bytranslistmean, expdf, verbose = FALSE) resauc <- allauc(bytranslistmean, expdf, nbwindows, verbose = FALSE) ## Testing attenuation resatt <- attenuation(resauc, resknee, rescountna, bytranslistmean, expdf, resmeandiff, verbose = FALSE) ``` ## See Also [allauc()], [kneeid()], [countna()], [meandifference()]

attenuation function