R: estimate Abundances from sonicated samples

estAbund {sonicLength}

R Documentation

estimate Abundances from sonicated samples

Description

When one or more replicate samples are sonicated and the lengths for each integration site are recorded, it is of interest to estimate actual number of sonicants allowing for coincidentally equal lengths. A likelihood approach is used here.

Usage

estAbund(locations, lengths, replicates = NULL, jackknife = F, kmax=0,
min.length = 20, ...)

Arguments

`locations`	a vector of IDs for distinct locations
`lengths`	a vector of the corresponding lengths
`replicates`	an optional vector of the replicate sample ID
`jackknife`	Whether to do leave-one-out jackknife samples over the replicates
`kmax`	highest count to bother with when computing mass function for counts. All higher values are globbed together in the result. If `kmax==0` don't compute
`min.length`	Least length expected for a fragment. Shorter lengths will not be permitted
`...`	Other arguments that may be passed along to functions that do the real work.

Details

This is a wrapper function for maxEM so study that function to see what is going on.

Value

a list with components

`theta`	estimated abundances
`phi`	estimated bin probabilities
`var.theta`	variances of `theta`
`iter`	number of iterations till convergence was achieved
`call`	the function call
`lframe`	the `data.frame` used to construct the estimate of `phi`
`obs`	the observed counts of distinct lengths
`jackknife`	a list of calls to `maxEM` with one replicate omitted in each
`data`	a `data.frame` with columns `locations`,`lengths`, and `replicates` (if supplied)

Author(s)

Charles C Berry ccberry@users.r-forge.r-project.org

References

Estimating abundances of retroviral insertion sites from DNA fragment length data. Berry CC, Gillet NA, Melamed A, Gormley N, Bangham CR, Bushman FD. Bioinformatics; 2012 Mar 15;28(6):755-62.