Comparison of different interpretation strategies for low template DNA mixtures.

DNA typing protocols have been improved over the last few years and even mixtures from minute and low grade DNA stains do not necessarily preclude typing. Nevertheless, in those cases stochastic phenomena tend to hamper interpretation. In order to supplement the current discussion about the interpretation of such challenging data, we focused on different combinations of analyses as an attempt to overcome stochastic problems. We analyzed mixtures of two types of degraded DNA in low template amounts (50 and 100pg DNA per contributor) using four types of multiplex STR typing kits: PowerPlex(®) ESX 17, PowerPlex(®) ESI 17, Investigator(®) ESSplex SE Kit and P11, a non-commercial kit. We employed the results of double or triple analyses for a comparison of different types of interpretation rules based on reporting either only reproducible alleles (consensus interpretation) or all alleles, even if they are only observed once (composite interpretation). The interpreted and composed profiles were compared to the known alleles of the contributors and a "degree of validity" was calculated. When only two single amplifications were taken into account, we observed a higher degree of validity for composite profiles. The difference for consensus interpretation could be compensated when a minimum of three amplifications were carried out. Using the same kit for repeat analyses increases the chances to yield reproducible results required for consensus interpretation. Combining different kits in a complementing approach, on the other hand, offers the opportunity to reduce the number of drop-out alleles: differences in amplicon lengths for specific markers between kits can increase the resulting information. In the case of a few amplifications available this effect might only be visible with the composite method. Several markers like SE33 are particularly affected by this. Based on our observations the consensus interpretation method may not reflect the original profile in an optimal way in some special cases like low template, degraded mixture stains. In those cases the composite interpretation could yield more complete results. However, such a composite profile should be used with caution and only for limited purposes. Generally recommending the consensus interpretation thus seems not to be justified: a more differentiated approach appears to be worthwhile, e.g. the amount of drop-outs, the number of replicates, choice and combination of kits and even a marker specific procedure might be taken into account.