prep_profile_dt: applies normalizations and transformations to prof_dt, a tidy...

In jrboyd/seqtsne: Application of t-sne to NGS Genomic Signal Profiles

applies normalizations and transformations to prof_dt, a tidy data.table of profiles.

Description

applies normalizations and transformations to prof_dt, a tidy data.table of profiles.

Usage

prep_profile_dt(prof_dt, norm_dt, qgr, cap_value = Inf, high_on_right = TRUE)

Arguments

prof_dt	data.table of ChIP-seq signal profiles.
norm_dt	data.table containing norm_factor values for tall_var/wide_var combinations.
qgr	GRanges
cap_value	numeric, ChIP-seq data is prone to outliers, which will wash out weaker signal if not properly capped. Default is Inf, i.e. no capping.
high_on_right	boolean, should profiles be flipped when most signal is on the left? Default is TRUE. This is appropriate when there is no biological significance to the orientation of a signal, i.e. mirror image profiles are equivalent.

Value

list of two items. prepared version of prof_dt and query_gr modified to reflect any flipping required by high_on_right.

Examples

data(profile_dt)
data(query_gr)
#typically, norm_dt is the same configuration table used to fetch prof_dt
#here we derive a new norm_dt that will reduce H3K4me3 to 30% of H3K27me3.
norm_dt = unique(profile_dt[, list(tall_var, wide_var)])
norm_dt[, norm_factor := ifelse(wide_var == "H3K4me3", .3, 1)]
prep_profile_dt(profile_dt, norm_dt, query_gr)

jrboyd/seqtsne documentation built on Nov. 5, 2022, 6:37 a.m.