Improving human forensics through advances in genetics, genomics and molecular biology

Forensic DNA profiling based on short tandem repeat (STR) markers currently allows the identification of persons already known to the investigating authorities. This technology has recently been improved in terms of the ability to analyse degraded DNA and low amounts of DNA, and has increased discrimination power.There are still some technical limitations to STR profiling, some of which can be overcome by using SNPs, which are more suitable for dealing with highly degraded DNA. Although the existing STR-based forensic DNA databases make it unlikely that SNPs will replace STRs for universal identification, SNPs are likely to improve human identification in disaster victim identification and in kinship analysis.The limitation of currently used Y-chromosomal STRs (Y-STRs) for male identification from male/female mixed samples is that it only allows the identification of groups of paternally related males. This can be overcome by applying rapidly mutating Y-STRs that provide male relative differentiation in many cases, allowing individual male identification via Y chromosome analysis.STR or SNP profiling can only identify persons previously known to the investigating authorities, a limitation that could be solved by forensic DNA phenotyping; that is, the inference of externally visible traits and biogeographic ancestry from crime scene DNA to provide intelligence leads for a police investigation that is searching for unknown persons.DNA-based biogeographic ancestry inferences are already possible on the level of at least larger geographic regions such as continents, and partly on subregional levels, using suitable SNPs that also allow the reconstruction of mixed ancestry. DNA-based inferences with a resolution of single countries are unlikely to ever become available.Current DNA-based appearance prediction includes group-specific traits such as eye colour, hair colour and age with categorical prediction accuracies suitable for practical applications, and additional group-specific traits such as skin colour, hair morphology or baldness may follow. Individual-specific DNA-based facial morphology prediction would be most appreciated for finding unknown persons, but is currently beyond our level of genetic knowledge.mRNA-based determination of the cellular origin of a crime scene sample, as is now possible for most relevant tissues in forensic practice, including skin, provides more accuracy than previously used presumptive methods. The use of DNA methylation markers for this purpose appears promising.Estimating the deposition time of a crime scene blood sample by using circadian biomarkers has become possible, although more markers are needed for detailed timing. Furthermore, estimating sample age based on differential RNA degradation appears promising.Future prospects include using next-generation (PCR-free) sequencing technologies to improve human identification in heavily degraded and mixed samples, more detailed DNA reconstruction of human appearance to allow stringent concentration of police investigation in the search for unknown suspects, and more detailed molecular approaches for crime scene reconstruction, such as in sample deposition timing and perhaps in the reconstruction of the physiological conditions of victims and perpetrators during criminal acts.