An alternative method would be to split a genome into a set of overlapping bins of variable sizes.
Villages are of variable size.
The spectral region between 4.7 and 0.5 p.p.m. of each NMR spectrum was binned in 21 variable size buckets corresponding to known metabolites.
Equal-read bins with variable sizes (6,000, 20,000, and 50,000) were created based on simulated reads from the hg19 reference genome (Baslan et al., 2012; Navin and Hicks, 2011) with median bin sizes of 458 kb, 137 kb, and 54 kb, respectively.
To identify instances of AD and LD, we first divided the genome into 5,000 variable size "bins" with an average size of ∼600 kb, following established methods (Evrony et al., 2012; Navin et al., 2011 ).
In applying ID, we discretized the variables: for example, the phenotype of high-temperature growth was discretized into four bins of equal size.
The input sequence(s) is divided evenly into sequence bins of size determined by the input bin size.
DE transcripts were binned based on GO or PFAM annotated function, yielding bins of differing size, n.
Scaffolds smaller than 100 kb were binned into one bin, scaffolds between 100 kb and 1 Mb were binned into two bins of equal size, scaffolds between 1 Mb and 1.5 Mb were divided into three bins, scaffolds between 1.5 and 2 Mb were divided into four bins, and so on.
