non_redundant_hier: Non Redundant set of samples using hierarchical search.

In irycisBioinfo/PATO: Pangenome Analysis Toolkit

Non Redundant set of samples using hierarchical search.

Description

Creates a non-redundant sub-set of samples. The function accept an accnet or mash object and returns a nr_list object. The sub-set can be created using a certain distance (mash o jaccard distance), a specific number of elements or a fraction of the whole sample set. The function performes an iterative seach so sometimes the exact number of returned elements could be not the same that the specified in the input. This difference can be defined with the threshold parameter. This function perform a hierarchical search. For small datasets (<2000) non_redundant() could be faster. This function comsume less memory than the original so it is better for large datasets.

Usage

non_redundant_hier(
  data,
  number,
  fraction,
  distance,
  tolerance = 0.05,
  partitions = 10,
  max_iter = 10000,
  fast = FALSE,
  snps
)

Arguments

data	An accnet or mash object.
number	The number of non-redundant samples.
fraction	The fraction of the whole set of non-redundant samples.
distance	Minimun distance among samples.
tolerance	Percentage of error between the input number and the final number of samples.
partitions	Number of partitions to hiaerarchical search.
max_iter	Maximun number of search iterations.
fast	If fast is TRUE the clustering process uses "components" in other case use "fast_greddy".

Value

nr_list object

See Also

extract_non_redundant

irycisBioinfo/PATO documentation built on Oct. 19, 2023, 3:07 p.m.