Short tandem repeats, or STRs, are stretches of DNA in which a short sequence of two to six base pairs is repeated over and over. The number of repeats at a given locus varies widely between individuals, which is precisely what makes these markers so valuable in forensic science. Because a person inherits one copy of each autosomal locus from each parent, a profile across many independent STR loci becomes a molecular fingerprint so rare that the chance of two unrelated people sharing it is vanishingly small. But that claim is only as strong as the population data behind it. Allele frequencies differ among human populations, so calculating the probability of a random match requires knowing how common each repeat variant is in the relevant population. Without local frequency data, forensic statisticians must borrow figures from other groups, introducing uncertainty that can undermine confidence in courtrooms.

Rwanda’s forensic DNA capability has grown rapidly in recent years, with genetic evidence now routinely used in criminal investigations, paternity disputes, and civil cases. Yet comprehensive population-specific reference data had lagged behind. The first forensic STR study in the country analyzed a relatively small cohort of unrelated individuals with a limited marker panel, offering an initial allele frequency dataset with restricted coverage. Earlier work, including a 2004 study of allele distribution among Rwandan Tutsi and a 2003 analysis of 16 STR loci in Hutu individuals, provided valuable but narrow snapshots. The new study dramatically expands that foundation, both in sample size and in the number of markers characterized.

The researchers took a retrospective approach, drawing on archived STR genotype data generated between 2005 and 2015. Through database sampling, they retrieved all 815 profiles from unrelated individuals that met the study’s inclusion criteria. Because the material spanned a full decade of laboratory work, the team painstakingly reviewed laboratory records to verify the extraction and quantification methods, amplification systems, capillary electrophoresis platforms, allele-calling software, and quality assurance procedures used throughout the period. This methodological audit ensured that data generated under different protocols over the years could be combined coherently into a single reference dataset.

The laboratory workflow itself reflects standard forensic practice of the era. DNA was extracted using the Chelex 100 method, a resin-based technique that binds metal ions and inhibiting contaminants while releasing template DNA. For casework samples, quantification followed with the Quantifiler Duo DNA Quantification kit on an ABI 7500 Real-Time PCR System, allowing technicians to confirm both the quantity of human DNA and the presence of inhibitors before amplification. The STR amplification combined the PowerPlex 16 system with PowerPlex ESI 17 Pro and PowerPlex ESX 17 kits, yielding a combined panel of 23 autosomal STR loci, including the highly discriminating SE33 marker that is standard in European forensic practice.

The results confirm that all 23 loci are robustly polymorphic in the Rwandan population, but the degree of variation varies considerably from marker to marker. The number of observed alleles per locus ranged from just 7 at D16S539 to a remarkable 50 at SE33, one of the most variable STR loci in the human genome. At most loci, one or two alleles predominated while the remaining variants appeared at relatively low frequencies. Among the most common were allele 16 at D3S1358, with a frequency of approximately 0.339; allele 7 at TH01, at roughly 0.378; allele 12 at D13S317, at about 0.363; allele 10 at D7S820, at around 0.406; and allele 12 at D5S818, at approximately 0.368. These patterns echo those seen in other Bantu-speaking populations of sub-Saharan Africa, consistent with Rwanda’s demographic history, while also revealing alleles rare enough elsewhere to be locally informative.

The headline statistic of the study is the combined match probability across the 23-locus panel: 1.7239 times 10 to the power of minus 30. In practical terms, if two profiles match at all 23 loci, the chance that a randomly selected unrelated Rwandan individual would share that same profile is roughly one in a nonillion, a number so extreme that it effectively removes any plausible ambiguity about identity for unrelated individuals. This extraordinarily low figure reflects the high informativeness of the combined panel, driven especially by hyper-variable loci such as SE33. It also means that even partial profiles recovered from degraded crime scene samples, where only a subset of loci amplifies successfully, can still carry enormous evidential weight when interpreted against the new frequency data.

The forensic value of the dataset extends beyond match probabilities. Allele frequencies feed into every major statistical framework used in DNA interpretation, including likelihood ratios, paternity indices, and kinship analyses. In paternity testing, for example, the strength of evidence for or against fatherhood depends on how common the child’s paternal alleles are in the population; a rare allele shared between alleged father and child is far more persuasive than a common one. Similarly, in disaster victim identification and missing persons investigations, accurate frequency estimates are essential for weighing the possibility of coincidental matches among relatives. By providing nationally distributed data, the study reduces the geographic and ethnic sampling bias that plagued earlier, more localized efforts.

The work also carries scientific significance beyond the courtroom. Rwanda occupies a key position in studies of East African population history, and its STR variation contributes to a broader picture of genetic diversity in sub-Saharan Africa, the region with the deepest human genetic diversity on Earth. Recent whole-genome sequencing efforts across 44 indigenous African populations have underscored how undersampled much of the continent remains in genetic databases. Expanded forensic datasets like this one, together with earlier mitochondrial DNA studies covering Côte d’Ivoire and Rwanda, help fill critical gaps that affect both forensic statistics and population genetics research. The detection of rare alleles in the Rwandan panel adds to the growing catalog of global STR diversity and improves the precision of profile probability estimates not only locally but in international databases that incorporate African frequency data.

For Rwanda, the immediate implications are practical. The Rwanda Forensic Institute, the Rwanda National Police, and the National Public Prosecution Authority, all partners in the research, now have a defensible, population-specific statistical foundation for DNA testimony. As DNA evidence becomes more central to the justice system, courts will increasingly demand that match statistics rest on frequencies measured in the relevant population rather than approximations from distant groups. The study, funded by the University of Rwanda and the European Union Team Europe Initiative under the Kwigira Project, also represents a model of South–North scientific collaboration, pairing Rwandan institutions with forensic specialists in Hamburg. With the expanded characterization of highly polymorphic loci and the detection of rare alleles, the authors conclude that the findings strengthen the statistical basis of forensic DNA interpretation in Rwanda and consolidate the country’s forensic genetic resources for years to come.

Subject of Research: Allele frequencies and forensic efficiency of autosomal STR loci in the Rwandan population

Article Title: Allele frequencies and forensic efficiency of autosomal short tandem repeat loci in the Rwandan population

Article References: Ndungutse, A., Daba, T. M., Krebs, O., Augustin, C., & Mutesa, L. (2026). Allele frequencies and forensic efficiency of autosomal short tandem repeat loci in the Rwandan population. International Journal of Legal Medicine. https://doi.org/10.1007/s00414-026-04022-6

Image Credits: AI Generated

DOI: 10.1007/s00414-026-04022-6

Keywords: forensic genetics, short tandem repeats, allele frequencies, Rwanda, DNA profiling, human identification, kinship analysis, polymerase chain reaction, combined match probability, International Journal of Legal Medicine, population genetics, forensic DNA

Short tandem repeats are stretches of DNA in which a short sequence of two to six base pairs is repeated over and over, such as ATATATAT. Because the number of repeats varies widely between individuals, STRs form the backbone of DNA profiling in human forensics. Standardized human STR genotyping systems, built on carefully validated panels of markers, allow laboratories around the world to produce comparable, court-admissible profiles. Animal forensics has never enjoyed that level of coordination. Validated STR markers for most domestic and wild species are scarce, and the markers that do exist are often dinucleotide STRs, repeats of just two base pairs, which are notoriously prone to genotyping artifacts such as stutter, the generation of spurious off-by-one peaks that complicate interpretation. Population data, which allow forensic scientists to calculate the statistical weight of a match, are frequently missing altogether.

The team behind pSTRminer, led by Jiajun Liu, Zhentang Liu, and senior authors Hongyu Sun and Riga Wu of the Faculty of Forensic Medicine at Zhongshan School of Medicine, set out to close these gaps with a single, scalable computational framework. The software automates what has traditionally been a fragmented, largely manual workflow: scanning a reference genome for STR loci, genotyping those loci in large collections of whole-genome sequencing data, and then scoring each locus for the properties that matter in forensic practice, including genotyping success rate and polymorphism information content, a standard measure of how informative a genetic marker is for distinguishing individuals.

To demonstrate the power of the approach, the researchers turned to domestic cattle, Bos taurus, one of the most economically and forensically significant livestock species in the world. Applying pSTRminer to the cattle reference genome, they identified 775,444 STRs de novo, a catalog of repeat loci far exceeding anything previously assembled for the species. They then genotyped this catalog using whole-genome sequencing data from 60 Chinese cattle and 111 African cattle, two populations chosen to represent sharply divergent genetic backgrounds. The logic is straightforward but important: a marker that appears highly variable in only one breed or region may be nearly useless elsewhere, so evaluating polymorphism across diverse lineages is essential before any locus can be recommended for global forensic use.

From this population-scale analysis, the team constructed the cattle STR database, or CSDB, a curated resource containing only those loci that met stringent quality criteria: a genotyping success rate of at least 40 percent and a polymorphism information content of at least 0.5. These thresholds ensure that the database holds markers that both amplify reliably in the laboratory and carry enough variation to discriminate between individuals. The sensitivity of the database to the genotyping success rate threshold was examined in supplementary analyses, giving future users a transparent view of how the marker set changes as criteria are tightened or relaxed.

Computational screening alone, however, is not enough for forensic work. Markers destined for casework must survive contact with real samples. The researchers therefore experimentally validated a panel of loci in a local Chinese cattle population of 145 animals using next-generation sequencing. Thirty tetranucleotide STRs, repeats of four base pairs, and 33 dinucleotide STRs were randomly selected from the database and tested. The validation confirmed that the markers were reliable, and it produced a nuanced picture of the trade-offs between repeat types. Tetranucleotide STRs showed lower average polymorphism than their dinucleotide counterparts, meaning they tend to be somewhat less variable across individuals. But they carried a decisive advantage: significantly lower stutter ratios, a difference the authors report as statistically significant at p less than 0.05. In practical terms, four-base-pair repeats generate fewer genotyping artifacts, producing cleaner, easier-to-interpret profiles.

That finding matters because stutter is one of the most persistent headaches in STR analysis. When a polymerase copies a repeat tract, it occasionally slips, adding or dropping a repeat unit and creating a minor artifact peak one repeat shorter or longer than the true allele. Dinucleotide repeats, with their short two-base motif, are especially vulnerable to this slippage. A marker with high stutter can obscure genuine alleles, particularly in degraded or low-template samples common in forensic contexts. By demonstrating that certain tetranucleotide STRs can actually surpass dinucleotide STRs in polymorphism while producing far fewer artifacts, the study lays out a viable path toward building animal STR panels that are both highly discriminative and technically robust. Systematic screening across the CSDB revealed that such high-performing tetranucleotide loci are not rare exceptions but a discoverable resource waiting to be tapped.

The broader significance of pSTRminer extends well beyond cattle. The software integrates established components of the modern genomics pipeline, drawing on widely used tools for read preprocessing, alignment, and STR genotyping, and wraps them into a reproducible workflow that reduces the manual operations required to move from raw sequencing data to a validated marker panel. The authors provide detailed documentation of the commands needed to reproduce their analyses, and supplementary tables include the formulas used to calculate forensic parameters, the overlap between STRs currently in use and those newly identified in cattle, and recommended analytical thresholds for heterozygote balance at validated loci. In effect, the study offers not just a database but a blueprint that other laboratories can follow to develop standardized STR systems for dogs, cats, horses, yaks, wildlife species, or any organism with a reference genome and population sequencing data.

The need for such tools is well documented in the forensic literature. Individual identification systems based on STR panels have been developed for domestic cats, dogs, and horses, and microsatellite marker sets have been proposed for parentage testing in cattle, yaks, and Chinese Holstein bulls. Yet each of these efforts has relied on comparatively small collections of markers, often selected without genome-wide polymorphism data or population-scale validation. Human forensics, by contrast, has moved decisively toward expanded multiplex systems and sequencing-based genotyping, supported by open population databases built from large-scale sequencing projects such as the 1000 Genomes Project. pSTRminer aims to bring animal forensics closer to that standard, enabling marker discovery at the scale the human field now takes for granted.

The work, supported by the National Natural Science Foundation of China, was approved by the Institutional Animal Care and Use Committee of Sun Yat-sen University, and the authors acknowledge the publicly available whole-genome sequencing data from the NCBI Sequence Read Archive that made the population analyses possible. For forensic scientists, the arrival of pSTRminer and the CSDB marks a shift from ad hoc marker selection to systematic, data-driven panel design. A single hair from a stolen calf, a bloodstain on a suspect’s boot, or a trace of tissue from a poached animal could soon be profiled with the same rigor and statistical confidence that human DNA evidence enjoys. As genome sequencing becomes cheaper and reference genomes accumulate for more species, the framework promises to make animal forensic genetics faster, cleaner, and more defensible, one well-validated repeat at a time.

Subject of Research: Genome-wide identification and population-scale evaluation of polymorphic short tandem repeats for animal forensic genetics

Article Title: pSTRminer: integrated bioinformatic software for genome-wide identification and population-scale evaluation of polymorphic short tandem repeats

Article References: pSTRminer: integrated bioinformatic software for genome-wide identification and population-scale evaluation of polymorphic short tandem repeats. (n.d.). https://doi.org/10.1007/s00414-026-04002-w

Image Credits: AI Generated

DOI: 10.1007/s00414-026-04002-w

Keywords: pSTRminer, short tandem repeats, forensic genetics, cattle, bioinformatics, genotyping, polymorphism, next-generation sequencing, STR database, animal forensics, population genetics, tetranucleotide STRs