But none of his convictions were for crimes so serious that he would have been required to submit a DNA sample, so the authorities could not compare his genetic code to evidence recovered in unsolved crimes.
In the MIL algorithm proposed by Babenko et al. [14], it needs to maximize Eq. 13, and this would compute additional M probabilities belonging positive or negative set for each sample, so the computing complexity is very high.
However, this Spanish validation has been studied in a multi-age sample; so, the MSBS seems to work in populations of any age.
We did expand our work to include a general population sample following exploratory factor analysis, but neither of our studies was carried out in a representative sample, so the mean scores may not reflect those in the general population.
Of importance, the sensitivities from Reference 13 were based on a single stool sample, so the relative change in sensitivity when increasing to four times the fecal matter can be expected to be similar.
One limitation of the earlier study is that it was based on a clinical sample, so the gender difference may have reflected referral bias (i.e., referred girls on average may have had a more severe phenotype).
