Run CLUMPP to aggregate multiple STRUCTURE runs.
clumpp(sr, k, align.algorithm = "greedy", sim.stat = "g",
greedy.option = "ran.order", repeats = 100, order.by.run = 0,
label = NULL, delete.files = TRUE)result from structure or folder name containing
STRUCTURE output files.
choice of k in sr to combine.
algorithm to be used for aligning the runs. Can be "full.search", "greedy", or "large.k".
pairwise matrix similarity statistic to be used. Can be "g" or "g.prime".
input order of runs to be tested. Required if
align.algorithm is "greedy" or "large.k". Valid choices are:
all |
test all possible input orders of runs (note that this option increases the run-time sub-stantially unless R is small). |
ran.order |
test a specified number of random input orders of
runs set by the repeats parameter. |
the number of input orders of runs to be tested. Only used if
align.algorithm is "greedy" or "large.k", and greedy.option
is "ran.order".
permute the clusters according to the cluster order of
a specific run. Set this parameter to a number from 1 to the number of
runs in sr.
label to use for input and output files.
logical. Delete all files when CLUMPP is finished?
Mattias Jakobsson and Noah A. Rosenberg. 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23(14):1801-1806. Available at http://web.stanford.edu/group/rosenberglab/clumppDownload.html