power.seqDepth: Estimate the marginal power under different sequencing depth
In PROPER: PROspective Power Evaluation for RNAseq

Description Usage Arguments Details Value Author(s) See Also Examples

The function helps study the power under different sequencing depth and sample sizes. It estimates the marginal powers based on existing simulation results, given new sequencing depth.

1	power.seqDepth(simResult, powerOutput, depth.factor = c(0.2, 0.5, 1, 2, 5, 10))

`simResult`	Result object from "runSims" function.
`powerOutput`	Result object from "comparePower" function.
`depth.factor`	A vector of numbers specifying the relative sequencing depth, comparing to the depth used in the simulation. 1 means using the same number of total reads as the simulation.

The powers under different sequencing depth and sample sizes provides important guidence in experimental design. Under the same total number of sequence reads, investigator can choose to use more replicates and shallower coverage for each, or less replicates and deeper coverage.

This function provides estimated marginal power holding all experimental variables fixed (biological variation, effect sizes, sample sizes, etc.) except the sequencing depth. Changing sequencing depth will only alter the marginal distribution of average counts. Since the stratified power (by average counts) won't change, those numbers are used in estimating the powers under different depth. This approaches allows skipping new simulations, which saves computation.

A matrix for marginal powers. Each row is for a sequencing depth, each columns is for a sample size.

Hao Wu <hao.wu@emory.edu>

comparePower, summary.power

## Not run: 
simOptions = RNAseq.SimOptions.2grp()
simRes = runSims(Nreps=c(3,5,7), sim.opts=simOptions, nsims=5,
                 DEmethod="edgeR")
powers = comparePower(simRes)
power.seqDepth(simRes, powers)

## End(Not run)