findParams: Finds the optimal values of parameters a and b that model the...

In krishan57gupta/ROSeq: Modeling expression ranks for noise-tolerant differential expression analysis of scRNA-Seq data

Description

Takes in as input the read count data corresponding to one sub-population and the typical gene statistics. Then it splits the entire range into equally sized bins of size k * σ, where k is a scalar with a default value of 0.05, and σ is the standard deviation of the pulled expression estimates across the cell-groups. Each of these bins corresponds to a rank. Therefore, for each group, cell frequency for each bin maps to a rank. These frequencies are normalized group-wise by dividing by the total cell count within a concerned group.

Usage

findParams(ds, geneStats)

Arguments

ds	The (normalized and filtered) read count data corresponding to a sub-population
geneStats	A vector containing 7 values corresponding to the gene data (maximum, minimum, mean, standard deviation, upper multiple of the standard deviation, lower multiple of standard deviation and log_2(fold change))

Value

results A vector containing 5 values (a, b, A, number of bins, R2)

krishan57gupta/ROSeq documentation built on March 3, 2021, 11:12 a.m.