The generalisability of artificial neural networks used to classify electrophoretic data produced under different conditions.

摘要: Abstract Previous work has shown that artificial neural networks can be used to classify signal in an electropherogram into categories have interpretational meaning (such as allele, baseline, pull-up or stutter). The previous trained the on a single data type, produced under laboratory condition and applied it was matched these factors. In this we investigate ability of different types (i.e. sourced profiles mixed DNA profiles) from conditions (specifically model electrophoresis instrument) determine whether set is required for each type network broad range still achieve same level performance. results our study implications how would choose train apply electropherograms their laboratory.