Figure 1. The method of multilocus DNA fingerprinting. After preparation from peripheral blood or any tissue, DNA is digested with an appropriate restriction enzyme, separated by gel electrophoresis, immobilized and hybridized to a minisatellite probe. Targeted minisatellites are specifically recognized by the probe molecules on the basis of complementarity; nonbound, excess probe is removed by washing. Signal detection (exposure to X-ray film using radioactively labeled probes or respective staining procedures for other recognition methods), documentation and evaluation follow (see Figure 2 for results and interpretations). The restriction enzyme/probe combination is chosen so as to optimize the information obtained from the DNA fingerprint patterns. For simplification, only two alleles from one minisatellite locus are covered in the top portion of the figure and highlighted by hatching in the DNA fingerprint pattern at the bottom. M refers to a marker lane that is always included in each DNA fingerprint gel in addition to length markers and as a methodological control.

Figure 2. Multilocus DNA fingerprinting for relationship analyses, for example in humans and animals (or plants). The putative parents, that is, the mother () and father (), and all of the offspring show unique individual DNA fingerprints. Seven of the offspring have a social and genetic father in common. The patterns of the remaining children prove their descent from another father (three sons in whose patterns the maternal contribution is clear, but the paternally derived bands do not comply with the suggested male). Due to the many alleles and the limited resolution of conventional gel systems, one to a few fragment size positions in a complex DNA fingerprint can also be common for unrelated individuals like the parents. Therefore, the origin of such a band cannot be traced back to either the mother or the father or both. The hypervariable length distributions of the many minisatellite alleles are obvious, but no bands can be identified as being allelic to each other. Thus data protection is ‘built-in’ in classical DNA fingerprints, since genetic linkage to any genes of interest cannot be deduced.

Figure 3. DNA profiling for relationship analyses. The putative parents, that is, the mother () and father (□ ), and all of the children (below the branching line) show different DNA profiles. The rationale of DNA profiling includes the demonstration of individual (hyper)variable loci, the alleles of which can readily be identified. Except for mutations, all alleles can be traced back to either the mother or the father or both. Single novel alleles in offspring – not visible in parents – may be due to (rare) mutations or other than the indicated descendence (see black arrows). In the case of paternal homozygosity for a specific allele, the lack of this allele in the profile of a child (□ ) indicates either mutation or wrong paternity open arrow pointing to the pattern of the son on the right-hand side.