If SNP information was missing in an individual then the scoring routine imputed expected values based on sample allele frequency.
The numerical method based on sampling the frequency response [21] can be an option, however, such an approach yields to approximated results with a high level of residual ISI.
In our study, only one individual is taken from each accession and the estimation can be performed after adjustment for population size; and second, the estimator is based on sample homozygosity, which is computed from allele frequency as given in Equation 1.
In order to identify the loci with the highest levels of disparity, we defined a score (S in eq. (1)) based on the sample frequency of asymmetrical shift for each heterozygous genotype, and all loci above or equal to 75 were selected (see Methods on allelic disparity), leaving 13,198 loci (Additional file 2, Figure S2).
Results from LS possible at higher frequencies are based on sampling only the short intervals, and are not appropriate when it is not known whether a signal is time-translation independent.
Based on the sample frequencies of each SNP, we defined a score (S) denoting the propensity of each SNP to lose either the B or A allele: (1) S = 1 - n (t B B j ) n (t A A j ). 100 for n (t A A j ) > n (t B B j ) 1 - n (t A A j ) n (t B B j ). 100 for n (t A A j ) < n (t B B j ) We can apply and modify an arbitrary threshold via Score (S).
However, since the method is based on equal-sampling-frequency, its accuracy is even worse than the EFS method.
The working principle is based on frequency sampling: assuming a roll-off factor of the prototype filter or smaller, each subchannel overlaps only with the immediately neighboring subchannels.
The third step involved separating true from artifactual variants within each sample, based on the frequency of variants within samples (F ij).
The method is based on a frequency sampling filter (FSF) representation of the process.
