pipeline.ExaML.bootstrap.per.proc: Compute phylogeny with bootstrap values

Description Usage Arguments Value Examples

View source: R/bigphylo.pipeline.R

Description

Compute a maximum likelihood phylogeny with the ExaML program. Bootstrap values capture the uncertainty in the tree phylogeny and are computed as follows. Columns of the multiple sequence alignment are bootstrapped to generate a random sequence data set. The starting tree is randomly constructed. A single phylogeny is constructed, and referred to as a boostrap phylogeny. Based on a set of bootstrap phylogenies, the one with best likelihood is determined and designated as the final ExaML phylogeny. Node uncertainty is computed by calculating how often any particular subtree in the final ExaML phylogeny occurs in the set of boostrap phylogenies.

Usage

1
2
3
4
pipeline.ExaML.bootstrap.per.proc(indir, infile, outdir = indir,
  bs.from = 0, bs.n = 500, bs.to = bs.n, hpc.walltime = 1,
  hpc.q = NA, hpc.sys = cmd.hpcsys(), hpc.mem = "500mb",
  hpc.nproc = 1, verbose = 1)

Arguments

indir

Directory of the sequences in file infile

infile

File name to an DNAbin matrix. The file name must end in .R

outdir

Output directory for the bootstrap calculations

bs.from

Iteration number of bootstrap calculations. Defaults to 0.

bs.n

Total number of boostrap iterations. Defaults to 500.

bs.to

Final iteration number of this call to pipeline.ExaML.bootstrap.per.proc. Defaults to bs.n.

hpc.walltime

Walltime specification for PBS header.

hpc.q

Queue specification for PBS header.

hpc.sys

Server specification for PBS header.

hpc.mem

RAM specification for PBS header.

hpc.nproc

Number of processors specification for PBS header.

verbose

Flag to run function in verbose mode.

Value

NULL. Creates shell files in outdir, and attempts to submit those to a queuing system.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
require(big.phylo)
data(nz_h3n2)
#	H3N2 data is stored in 'seq' DNAbin matrix object
print( seq )							
#	write sequences to directory for processing
indir		<- getwd()
infile		<- 'nz_h3n2.R'
insignat	<- ''
save(seq, file=paste(indir, infile, sep='/'))	
#	Write the a shell file to create the first ExaML boostrap tree
pipeline.ExaML.bootstrap.per.proc(indir, infile, bs.from=0, bs.to=0, hpc.walltime=1, hpc.q=NA, hpc.mem="500mb", hpc.nproc=1)
#	Resume and write the shell files to create two more ExaML boostrap trees
pipeline.ExaML.bootstrap.per.proc(indir, infile, bs.from=1, bs.to=2, hpc.walltime=1, hpc.q=NA, hpc.mem="500mb", hpc.nproc=1)
#
#	NOTE: if the current Unix system does not have a 'qsub' command, then a shell file will be generated that can
#	be run manually on the current Unix system. Support for an HPC system is currently implemented for the Imperial
#	server 'cx1.hpc.ic.ac.uk'. To see how the final HPC script would look like, try
#
#	pipeline.ExaML.bootstrap.per.proc(indir, infile, bs.from=1, bs.to=2, hpc.sys='cx1.hpc.ic.ac.uk', hpc.walltime=1, hpc.q=NA, hpc.mem="500mb", hpc.nproc=1)
#
#	It is not difficult to enable support for a different server.
#	See ?cmd.hpcwrapper

olli0601/big.phylo documentation built on Nov. 9, 2019, 9:25 a.m.