permute_pam: Function to permute a Presence-Absence-Matrix.
Arguments
- m
Presence-Absence-Matrix (PAM) or a binary matrix with columns representing species and rows sites.
- niter
Number of iterations to permute the PAM.
- as_sparse
If TRUE the PAM will be returned as a sparse matrix
- randal
Randomization algorithm applied to the PAM. Possible choices "curveball" and "fastball".
Value
Returns a permuted matrix of the same dimensions of m (same number of rows and columns). Note that the sum of each row and column of this permuted matrix is equal to that of m. species.
Details
This function uses the "curveball" (Strona et al., 2014) and the fastball (Godard and Neal, 2022) algorithms. The implementation of the "fastball" in C++ is provided in https://github.com/zpneal/fastball/blob/main/fastball.cpp. Please when using the "fastball" algorithm for publications cite Godard and Neal (2022). When using the "curveball" cite Strona et al. (2014).
References
Strona G, Nappo D, Boccacci F, Fattorini S, San-Miguel-Ayanz J (2014). “A fast and unbiased procedure to randomize ecological binary matrices with fixed row and column totals.” Nature Communications, 5(1), 1--9. ISSN 20411723, doi:10.1038/ncomms5114 , https://www.r-project.org. . Godard K, Neal ZP (2022). “fastball: a fast algorithm to randomly sample bipartite graphs with fixed degree sequences.” Journal of Complex Networks, 10(6), cnac049. ISSN 2051-1329, doi:10.1093/comnet/cnac049 , https://academic.oup.com/comnet/article-pdf/10/6/cnac049/47758701/cnac049.pdf.
Examples
set.seed(111)
pam <- matrix(rbinom(100,1,0.3),nrow = 10,ncol = 10)
ppam <- bamm::permute_pam(m = pam,niter = NULL,as_sparse = FALSE)
# Check if matrices are different
all(pam == ppam)
#> [1] FALSE
# Check if row totals are the same
all(Matrix::rowSums(pam) == Matrix::rowSums(ppam))
#> [1] TRUE
# Check if column total are the same
all(Matrix::colSums(pam) == Matrix::colSums(ppam))
#> [1] TRUE