Population Genetic Data for 23 STR Loci of Tawahka Ethnic Group in Honduras

Abstract

Background: The Tawahka ethnic group, with approximately 2690 individuals in northeastern Honduras, represents one of the country’s smallest indigenous communities. No genetic studies have been published on this population, and population-specific databases are essential for forensic applications. Methods: Allele frequencies for 23 autosomal short tandem repeats (STRs) loci were analyzed in 100 unrelated Tawahka individuals (61 females, 39 males) from the municipality of Wampusirpi. Deoxyribonucleic acid (DNA) was extracted from blood on Fast Technology for Analysis of nucleic acids (FTA) cards and amplified using the PowerPlex Fusion 6C System. Statistical parameters were calculated using Genepop v4.2 and Arlequin v5.3.2.2. Results: All loci showed substantial polymorphism with no Hardy–Weinberg equilibrium deviations after Bonferroni correction (α = 0.0022). Expected heterozygosity ranged from 0.4968 to 0.8812. Combined power of discrimination was 99.9999% and combined chance of exclusion was 99.99%. Conclusions: This first genetic characterization of the Tawahka population provides essential reference data for forensic identification, paternity testing, and population genetics studies. The dataset contributes to understanding indigenous Central American genetic diversity and ensures accurate forensic analyses for individuals of Tawahka ancestry following Combined DNA Index System (CODIS) and European Standard Set (ESS) standards.

1. Introduction

Events involving human identification through genetic analysis have become fundamental in modern forensic science and judicial procedures. Short tandem repeats (STRs) constitute the gold standard for human identification due to their high polymorphism, reproducibility, and suitability for statistical analysis [1,2]. The establishment of population-specific allele frequency databases is essential for accurate likelihood ratio calculations in forensic casework and paternity testing [3,4].

The Tawahka people represent one of the smallest indigenous communities in Honduras, with a total population of approximately 2690 individuals in 2023, accounting for approximately 0.03% of the country’s population [5]. This population is primarily concentrated in northeastern Honduras, specifically in the departments of Olancho and Gracias a Dios [6,7] (see Figure 1). Despite their cultural and historical significance, no genetic studies have been published on this ethnic group to date.

Honduras is home to nine indigenous and Afro-descendant peoples, each with distinct cultural, linguistic, and potentially genetic characteristics [8]. Previous genetic studies have characterized other Honduran populations, including the Lenca ethnic group [9], but the Tawahka remain genetically uncharacterized. Understanding the genetic structure of indigenous populations is crucial not only for forensic applications but also for anthropological studies, conservation genetics, and medical genetics [10,11].

The PowerPlex Fusion 6C System (Promega Corporation, Madison, WI, USA) is a multiplex STR amplification kit that enables simultaneous amplification of 23 autosomal STR loci, three Y-STR markers, and the amelogenin gender marker [12]. This system has been widely adopted in forensic laboratories worldwide due to its high discrimination power, sensitivity, and compatibility with degraded DNA samples [13,14].

This study follows the methodology previously used to characterize the Lenca population in Honduras [9] and aims to: (1) establish the first genetic reference database for the Tawahka ethnic group; (2) calculate forensic statistical parameters for individual identification; (3) evaluate Hardy–Weinberg equilibrium across all loci; and (4) provide data for future population genetics and anthropological studies.

In Honduras, forensic genetic analysis is centralized at the Forensic Medicine Directorate of the Public Ministry, which processes approximately 800–1200 STR profiles annually for criminal investigations and paternity testing [15]. Currently, forensic statistical calculations rely on a composite reference database comprising three primary sources: (1) allele frequency data from the Honduran mestizo population (N = 200), established in 2018 using the PowerPlex Fusion System (Promega Corporation, Madison, WI, USA) [16]; (2) data from the Lenca indigenous population (N = 100), published in 2024 using PowerPlex Fusion 6C (Promega Corporation, Madison, WI, USA) [9]; and (3) generic “Hispanic” allele frequencies from the Federal Bureau of Investigation (FBI) CODIS database, used when Honduran data are insufficient [17].

Match probabilities (random match probability, RMP) are calculated using the product rule with theta correction (θ = 0.01) to account for potential population substructure [3,18]. Probabilities are derived according to established formulas for homozygotes and heterozygotes, where p_i and p_j denote allele frequencies and θ represents the coancestry coefficient. Paternity indices (PIs) are computed as the ratio between the probability of observing the child’s genotype under the hypothesis that the alleged father is the biological father and the probability of observing it under the hypothesis of an unrelated random man. Formula (1) is used for homozygotes and formula (2) for heterozygotes:

$$P\left(A_i,A_i\right)={\displaystyle \frac{\left[2\theta +\left(1-\theta \right)p_i\right]\left[3\theta +\left(1-\theta \right)p_i\right]}{\left(1+\theta \right)\left(1+2\theta \right)}}$$

(1)

$$P\left(A_i,A_j\right)={\displaystyle \frac{2\left[\theta +\left(1-\theta \right)p_i\right]\left[\theta +\left(1-\theta \right)p_j\right]}{\left(1+\theta \right)\left(1+2\theta \right)}},i\ne j$$

(2)

Database selection follows a hierarchical approach: when an individual’s ethnicity is known and a corresponding database exists (e.g., Lenca), allele frequencies from that population are applied; when ancestry is unknown or admixed, the Honduran mestizo database is used; and in conservative approaches, the database yielding the most favorable result for the defendant may be selected.

Prior to this study, STR allele frequency data were unavailable for six of Honduras’s nine indigenous and Afro-descendant groups (Tawahka, Pech, Tolupán, Chortí, Garifuna and Miskito). Consequently, forensic analyses involving individuals from these populations relied on the mestizo reference database, potentially introducing bias in likelihood ratio estimations. The Tawahka dataset presented in this study addresses this gap by providing population-specific allele frequencies for one of these underrepresented groups and establishing a framework for future population genetic studies across Honduras’s diverse ethnic communities.

2. Materials and Methods

2.1. Sample Collection

Samples were collected from 100 unrelated healthy individuals (61 females, 39 males) from four settlements: Krausirpi (n = 42), Yapuwas (n = 28), Brisas de Pisijire (n = 18), and Krautura (n = 12), all within the municipality of Wampusirpi, department of Gracias a Dios. The sex distribution reflects the demographic composition and voluntary participation rates in these communities. Informed consent was obtained from each participant prior to sample collection. Participants confirmed their Tawahka ancestry for at least three generations and had no known family relationships with other participants.

To ensure sample independence, we implemented three complementary strategies: (1) participants confirmed the absence of known first- or second-degree relatives among other participants through structured interviews; (2) we verified that no individuals shared identical 23-locus STR profiles; and (3) we calculated pairwise identity-by-state (IBS) values across all samples using the 23 autosomal STRs, confirming that no individual pairs exceeded the threshold indicative of first-degree relatedness (IBS > 0.50 for parent-offspring or full siblings). All pairwise IBS values ranged from 0.18 to 0.42 (mean = 0.28 ± 0.05), consistent with unrelated individuals from the same population [19].

2.2. DNA Extraction

DNA was extracted from 2 mL of blood collected using the venipuncture technique [20] and then impregnated onto Indicating FTA Cards (Qiagen, Germantown, MD, USA) [21]. FTA cards provide a convenient method for DNA collection, transport, and long-term storage at room temperature while protecting DNA from degradation [22,23].

2.3. PCR Amplification

Direct amplifications of Polymerase Chain Reaction (PCR) on the FTA cards were performed using the PowerPlex Fusion 6C™ System (Promega Corporation, Madison, WI, USA) according to the manufacturer’s recommendations [24]. This system amplifies 27 loci simultaneously: 23 autosomal STRs (CSF1PO, D1S1656, D2S441, D2S1338, D3S1358, D5S818, D7S820, D8S1179, D10S1248, D12S391, D13S317, D16S539, D18S51, D19S433, D21S11, D22S1045, FGA, Penta D, Penta E, SE33, TH01, TPOX, and vWA), three Y-STR loci (DYS391, DYS576, and DYS570), and the amelogenin gender marker.

2.4. Fragment Analysis and Genotyping

Amplified products were analyzed by capillary electrophoresis using an Applied Biosystems 3500 (ThemoFisher Scientific, Waltham, MA, USA) automated sequencer [25] and GeneMapper v1.4 software [26]. The designations of alleles adhered to the guidelines established by the DNA Commission of the International Society for Forensic Genetics (ISFG), utilizing allelic ladders supplied by the manufacturer [27,28].

2.5. Statistical Analysis

Allelic-frequency computations were performed in Genepop v4.2 [29] and Arlequin v5.3.2.2 [30]. For each locus, we estimated the power of discrimination (PD) and chance of exclusion (CE) following Huston’s approach [31], and derived polymorphic information content (PIC) as described by Botstein et al. [32]. Hardy–Weinberg equilibrium was evaluated using the GDA program [33] via exact tests, while Analysis of Molecular Variance (AMOVA) exact tests were implemented in Arlequin [34]. To control family-wise errors across loci, we applied a Bonferroni adjustment [35], yielding a significance threshold of α = 0.0022 (0.05/23) for all locus-specific tests.

We acknowledge that the Bonferroni correction (α = 0.0022 for 23 loci) represents a conservative approach that assumes complete independence among loci. While STRs on different chromosomes are generally unlinked, this strict correction may inflate Type II error rates (false negatives), potentially obscuring subtle deviations from Hardy–Weinberg equilibrium (HWE) that could signal population substructure, recent admixture, or null alleles [35,36]. With our sample size (N = 100), the statistical power to detect moderate departures from HWE is already limited; the Bonferroni adjustment further reduces sensitivity.

Despite these limitations, we retained the Bonferroni correction for three reasons: (1) it provides a stringent quality control benchmark consistent with forensic genetic standards; (2) it minimizes false-positive identification of problematic loci; and (3) our primary objective is to establish a reliable reference database rather than investigate fine-scale population structure. Alternative approaches such as the false discovery rate (FDR) or sequential Bonferroni methods could be considered in future studies with larger sample sizes focused on demographic inference [37,38].

2.6. Quality Control

All analyses were performed in accordance with GITAD (Ibero-American Working Group on DNA Analysis) procedures and controls to ensure reliability [39]. Each amplification batch incorporated positive and negative controls, and allele assignments were calibrated with an allelic ladder and internal lane standards to secure accurate sizing. We validated analytical thresholds and monitored stutter ratios to minimize calling artifacts. Laboratory performance was further verified through participation in inter-laboratory proficiency testing, collectively establishing robust quality assurance for the dataset.

2.7. Ethics Statement

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Biomedical Research Committee (CEIB) of the Scientific Research Unit (UIC), Faculty of Medical Sciences, National Autonomous University of Honduras (UNAH) (IRB-00003070). The study protocol adhered to all applicable national and international standards, including the Honduran National Bioethics Regulations, ISFG recommendations for population genetic studies, and principles outlined in the Nagoya Protocol on Access and Benefit-Sharing [40].

2.7.1. Informed Consent and Community Engagement

Prior to enrollment, participants received clear, culturally appropriate written and verbal information in Spanish (and Tawahka when possible, facilitated by community interpreters) about the study’s purpose, procedures, risks, benefits, voluntary nature of participation, and their right to withdraw at any time without consequence. Particular emphasis was placed on explaining: (1) how genetic data would be used for forensic and research purposes; (2) safeguards for confidentiality and security; (3) potential benefits to the Tawahka community through improved forensic services; and (4) the absence of direct individual medical benefits. All participants provided written informed consent. For participants with limited literacy, the consent process included oral explanation with witness signatures. Community leaders from each settlement were consulted throughout the study design and implementation phases to ensure cultural appropriateness and community acceptance.

2.7.2. Data Sovereignty and Benefit-Sharing

Recognizing the Tawahka people’s inherent rights over their genetic resources and associated traditional knowledge [41], we implemented the following measures to uphold data sovereignty principles:

• Community authorization: In addition to individual informed consent, we obtained authorization from the Tawahka Indigenous Federation (FITH) representing the community’s collective interests before initiating sample collection.
• Data access and control: The complete dataset, including individual genotypes, is held jointly by the Forensic Medicine Directorate of the Honduran Public Ministry and the Tawahka community leadership. External researchers seeking access to individual-level data must obtain permission from both entities. The aggregated allele frequency data (Supplementary File S1) are made publicly available to maximize forensic utility while protecting individual privacy.
• Benefit-sharing: Benefits accruing from this research include: (a) priority access to forensic genetic services for Tawahka individuals through the Honduran Public Ministry; (b) capacity-building workshops on forensic genetics for community members; (c) co-authorship opportunities for Tawahka representatives in future publications; and (d) financial support for community-directed projects from a portion of any commercial licensing fees, should the database be incorporated into commercial forensic kits.
• Cultural sensitivity: We acknowledge that genetic research in indigenous communities raises complex ethical issues regarding group identity, stigmatization, and potential misuse [42,43]. Throughout the study, we emphasized that genetic diversity data reflect evolutionary history rather than biological or cultural superiority. We committed to avoiding sensationalized language in publications and to consulting with community leaders regarding any findings before public dissemination.

2.7.3. Compliance with the Nagoya Protocol

Honduras ratified the Nagoya Protocol in 2015 [40]. Our study complies with Protocol requirements through: (1) obtaining prior informed consent from both individuals and the Tawahka community; (2) negotiating mutually agreed terms for benefit-sharing; (3) ensuring that the use of genetic resources remains confined to stated forensic and population genetics purposes; and (4) maintaining transparent communication with the Tawahka community regarding data use and dissemination.

We recognize that standards for indigenous genetic research continue to evolve [44]. We remain committed to ongoing dialog with the Tawahka community and adaptation of our practices to reflect best practices in ethical indigenous research.

3. Results

3.1. Dataset

The dataset comprises allele-frequency data for 23 autosomal STR loci generated from 100 unrelated Tawahka individuals and is summarized in Supplementary File S1. For each locus, we report the observed allele designations and their frequencies, alongside the effective sample size (N = 100), observed heterozygosity (Ho), and expected heterozygosity under Hardy–Weinberg assumptions (He). Forensic informativeness was quantified through the power of discrimination (PD), chance of exclusion (CE), and polymorphic information content (PIC), with the minimum allele frequency (MAF) provided to guide conservative statistical treatment of rare variants. Departures from Hardy–Weinberg equilibrium were evaluated using exact tests, and the corresponding p-values are listed to allow locus-specific quality control and interpretation. Together, these metrics afford a comprehensive view of locus variability and evidentiary value in this population, and they enable the computation of multilocus parameters (e.g., combined PD and cumulative CE) for downstream forensic applications.

Sex-stratified analysis of heterozygosity revealed no significant differences between males (mean Ho = 0.733 ± 0.104) and females (mean Ho = 0.741 ± 0.108) across the 23 autosomal loci (paired t-test: t = 0.62, p = 0.54), confirming the absence of sex-linked genotyping bias or allele dropout. Similarly, no locus showed significant heterozygosity differences between the four sampled settlements (ANOVA: all p > 0.15 after Bonferroni correction), indicating genetic homogeneity across this geographic region.

3.2. Geographic Context

The study population was sampled from four settlements within the primary Tawahka habitation area in northeastern Honduras: Krausirpi (15.05° N, 84.65° W; n = 42), Yapuwas (15.08° N, 84.72° W; n = 28), Brisas de Pisijire (15.12° N, 84.68° W; n = 18), and Krautura (15.09° N, 84.75° W; n = 12), all located within the municipality of Wampusirpi, department of Gracias a Dios (Figure 1). This region is characterized by tropical rainforest, limited road access, and geographic isolation that has historically preserved the cultural and linguistic distinctiveness of the Tawahka people [6,16].

Figure 1. Geographic distribution of the Tawahka ethnic group in Honduras. The yellow region delineates the primary habitation area of the Tawahka people in northeastern Honduras, spanning portions of the departments of Olancho and Gracias a Dios. The study population was sampled from four settlements within this region: Krausirpi (15.05° N, 84.65° W), Yapuwas (15.08° N, 84.72° W), Brisas de Pisijire (15.12° N, 84.68° W), and Krautura (15.09° N, 84.75° W), all located within the municipality of Wampusirpi. The region is characterized by tropical rainforest, limited road access, and geographic isolation that has historically preserved the cultural and linguistic distinctiveness of the Tawahka people [6,16]. Map modified from D-maps [45] and Native-Land [46].

Figure 1. Geographic distribution of the Tawahka ethnic group in Honduras. The yellow region delineates the primary habitation area of the Tawahka people in northeastern Honduras, spanning portions of the departments of Olancho and Gracias a Dios. The study population was sampled from four settlements within this region: Krausirpi (15.05° N, 84.65° W), Yapuwas (15.08° N, 84.72° W), Brisas de Pisijire (15.12° N, 84.68° W), and Krautura (15.09° N, 84.75° W), all located within the municipality of Wampusirpi. The region is characterized by tropical rainforest, limited road access, and geographic isolation that has historically preserved the cultural and linguistic distinctiveness of the Tawahka people [6,16]. Map modified from D-maps [45] and Native-Land [46].

3.3. Statistical Parameters

Key statistical indicators underscore the strong forensic performance of the 23-locus STR panel in the Tawahka sample. The combined power of discrimination (PD) calculated using the product rule with theta correction (θ = 0.01) was >99.9999% (combined RMP < 1 in 10¹⁴), implying an exceptionally small probability of observing a matching profile by chance in an unrelated individual. The combined chance of exclusion (CE) was >99.99% (1 − Q = 0.9999), supporting strong effectiveness for parentage testing.

These statistics assume linkage equilibrium (independence) among loci, an assumption supported by the fact that our 23 markers are distributed across 14 autosomes with no known physical linkage [12,14]. However, we acknowledge potential sources of multi-locus dependencies:

• Population substructure: If the Tawahka comprises genetically distinct subgroups, alleles at different loci could exhibit correlations (Wahlund effect) [47]. Our fixation index (FST) analysis across four settlements detected minimal structure (mean FST = 0.003), suggesting this effect is negligible.
• Admixture: If recent admixtures with neighboring populations have occurred, ancestry blocks could introduce transient linkage disequilibrium. Future genome-wide studies would clarify this issue.
• Cryptic relatedness: Although we screened for known relatives, distant coancestry could inflate homozygosity and generate correlations. The theta correction partially accounts for this.

Given these considerations, we report combined statistics with conservative rounding (>99.9999%) to avoid implying spurious precision. The precise RMP value depends on assumptions (θ, linkage equilibrium, database accuracy) that introduce uncertainty exceeding the nominal precision of 20+ significant figures. For practical forensic purposes, any RMP below 1 in 10¹² constitutes extremely strong evidence of identity [3,48], rendering additional decimal places forensically and legally irrelevant.

Tests of Hardy–Weinberg equilibrium revealed no significant deviations after Bonferroni correction (α = 0.0022), consistent with appropriate sampling and the absence of detectable inbreeding or substructure. Together, these results confirm the panel’s exceptional discriminatory capacity for individual identification and its suitability for kinship applications in this population.

3.4. Loci Characteristics

Across the 23 autosomal STRs, levels of polymorphism varied widely. A subset of loci showed high diversity (He > 0.80)—including D1S1656, D13S317, Penta E, D16S539, D18S51, D2S1338, Penta D, D8S1179, D12S391, D19S433, SE33, and FGA indicating strong informativeness for individualization. Moderate diversity (He = 0.60–0.80) characterized D3S1358, D10S1248, CSF1PO, TH01, vWA, D21S11, D7S820, D5S818, and TPOX, which still contribute substantially to multilocus power. Lower diversity (He < 0.60) was observed at D2S441 and D22S1045. Consistent with this pattern, SE33 was the most polymorphic marker (He = 0.8812), whereas D2S441 exhibited the least polymorphism (He = 0.4968).

3.5. Data Availability and Repository

The complete dataset accompanies this article as Supplementary Material and has also been deposited for open access. A mirrored copy can be obtained on request from the Forensic Medicine Directorate, Public Ministry of Honduras (https://www.mp.hn, accessed on 10 November 2025). Following the example of the National Institute of Standards and Technology (NIST) 1036 United States (USA) Population Database [49] and recent recommendations for transparent forensic data sharing [50,51], we provide complete 23-locus STR profiles for all 100 Tawahka individuals as Supplementary File S1 (Excel format) with in file description and data dictionary. All genotypes have been verified through duplicate amplification and independent review by two analysts, ensuring data quality suitable for reanalysis and meta-analyses.

3.5.1. Privacy and Ethical Safeguards

Although multilocus STR profiles are not considered directly identifying information under Honduran legislation [52], multiple layers of protection were implemented to preserve participant confidentiality and uphold ethical standards:

• De-identification: All personal identifiers (names, birthdates, addresses) were permanently removed. Sample identifiers are non-reversible pseudonyms, and the linkage key is securely stored and inaccessible to external parties.
• Aggregate release: The dataset is provided as a complete set of profiles to preclude individual singling-out.
• Community authorization: The Tawahka Indigenous Federation (FITH) reviewed and approved the release of individual-level genotypes, confirming that community concerns related to privacy, stigmatization, and potential misuse were addressed satisfactorily.
• Use restrictions: Data are openly available for research and forensic applications, provided that users (a) cite this publication, (b) refrain from re-identification attempts, (c) avoid discriminatory use, and (d) acknowledge the Tawahka community in derivative works.

3.5.2. Utility for Statistical Verification

The availability of complete multilocus profiles allows independent researchers to:

• Verify reported allele frequencies and population parameters.
• Recalculate forensic statistics under alternative assumptions (e.g., varying θ values).
• Explore population structure through FST, AMOVA, and Principal Component Analysis (PCA) analyses.
• Conduct simulations to evaluate exclusion and discrimination power.
• Compare Tawahka genotypes against other reference datasets (e.g., NIST, EDNAP (European DNA Profiling Group) Mitochondrial DNA Population Database (EMPOP)).
• Quantify inter-database biases in profile frequency estimation.

These data are intended to support methodological validation, algorithmic testing, and cross-population comparisons. Such transparency contributes to the advancement of forensic genetics and reinforces the evidentiary reliability of population-based statistical inference [50,51].

3.5.3. Profile Frequency Verification

To enable independent confirmation of our statistical estimates, random match probabilities (RMPs) were computed for all 100 Tawahka profiles using three databases: the Tawahka dataset (this study), the Honduran mestizo population [16], and the NIST Caucasian database [17]. The mean RMP values were as follows:

• Tawahka: 1 in 5.8 × 10¹⁴ (range: 1 in 8.3 × 10¹²–1 in 4.1 × 10¹⁶);
• Honduran mestizo: 1 in 2.3 × 10¹⁵ (range: 1 in 3.1 × 10¹³–1 in 1.5 × 10¹⁷);
• NIST Caucasian: 1 in 8.7 × 10¹⁵ (range: 1 in 1.1 × 10¹⁴–1 in 5.2 × 10¹⁷).

These results confirm that population-appropriate allele frequencies (Tawahka) produce more conservative RMP values, with mean fold differences of 4.0× and 15.0× relative to the Honduran mestizo and NIST Caucasian datasets, respectively. The broad range of RMP values reflects genuine population diversity, where certain genotypes are extremely rare (approaching 1 in 10¹⁶) and others more common due to high-frequency allele combinations.

When Tawahka genotypes were evaluated using the NIST Caucasian database, 12 profiles yielded RMPs exceeding 1 in 10¹⁷, highlighting the risk of overestimated evidentiary weight when using non-representative reference data. Conversely, calculating NIST Caucasian profile frequencies with the Tawahka dataset produced underestimates (mean fold difference: 0.18×), demonstrating reciprocal bias.

These findings emphasize the necessity of population-specific databases and substantiate the importance of releasing complete multilocus datasets for robust forensic interpretation and inter-population validation. All Supplementary Files S1–S3 include standardized field labels and detailed metadata. Users are required to cite this publication and acknowledge the Tawahka community when reusing these data.

4. Discussion

4.1. Alleles

Allele-frequency analysis indicated substantial genetic diversity in the Tawahka sample across all 23 STR loci. In total, 198 distinct alleles were observed, with allelic richness per locus ranging from 5 alleles (D2S441, TH01, TPOX) to 21 alleles (SE33), yielding an average of 8.61 ± 4.15 alleles per locus (mean ± SD). This level of diversity is comparable to other Central American indigenous populations: the Honduran Lenca exhibited a mean of 8.91 alleles per locus [9], Guatemalan Maya populations showed 7.2–9.5 alleles per locus depending on the marker set [53,54], and admixed Guatemalan Ladino populations displayed 9.1–10.8 alleles per locus [55].

Consistent with regional patterns, the pentanucleotide markers were among the most informative: Penta D (9 alleles; He = 0.8170) and Penta E (15 alleles; He = 0.8510) exhibited high heterozygosity, aligning with observations reported for the Lenca and other Central American populations [9]. The hypervariable SE33 locus showed the highest allelic diversity (21 alleles; He = 0.8812), while tetranucleotide markers generally displayed moderate diversity (5–11 alleles).

Allele distributions varied by locus: markers such as D16S539, D7S820, and D12S391 showed relatively balanced frequency spectra, whereas others displayed dominant alleles exceeding 40% frequency (e.g., D2S441 allele 10 at 66%; TPOX allele 8 at 34.5%; D22S1045 allele 16 at 60%). Notably, several rare variants (frequency < 0.01), including microvariants at D21S11, D19S433, and SE33, were detected, further enhancing the overall discriminatory capacity of the multilocus panel.

4.2. Hardy–Weinberg Equilibrium

Hardy–Weinberg equilibrium (HWE) tests showed no significant departures at any of the 23 loci after Bonferroni adjustment (α = 0.0022). Locus-specific p-values ranged from 0.0870 (D1S1656) to 0.9950 (D19S433), all comfortably above the corrected threshold, indicating that genotype frequencies conform to HWE expectations. This pattern is consistent with a sample of unrelated individuals drawn from a population in genetic equilibrium, with no detectable substructure, inbreeding, or recent admixture affecting these markers. Accordingly, the dataset is well suited for downstream forensic statistics, including unbiased estimation of random match probabilities and kinship likelihoods.

4.3. Forensic Statistical Parameters

The 23-locus STR panel exhibited outstanding forensic performance across identification and kinship contexts. For individualization, the combined power of discrimination reached 99.9999%, implying an exceedingly small residual match probability, with locus-specific PD values spanning 0.6698 (D2S441) to 0.9580 (SE33). In paternity testing, the combined chance of exclusion was 99.99%, supported by locus-level CE values from 0.1875 (D2S441) to 0.8776 (D1S1656), providing strong capacity to exclude non-fathers. Information content metrics were likewise high: PIC values ranged from 0.4330 (D2S441) to 0.8657 (SE33), with 18/23 loci exceeding 0.6. Collectively, the mixture of highly polymorphic markers yields robust statistical power for both casework identification and kinship inference.

4.4. Population Substructure and Theta Correction

The theta (θ) correction, also known as the coancestry coefficient or FST, accounts for potential non-independence of alleles due to population substructure, inbreeding, or genetic drift [18,56]. In forensic calculations, θ adjusts match probabilities to be more conservative (higher RMP, lower likelihood ratio) when individuals may be drawn from a subpopulated group rather than a panmictic population.

Following international recommendations [3,18,56], Honduran forensic laboratories currently apply a default θ = 0.01 for all calculations, regardless of population. This value represents a “generic” correction broadly appropriate for moderately structured populations. However, optimal θ values are population-specific and should ideally be estimated from empirical data [56,57]. We estimated FST for the Tawahka using two approaches:

• Within-population estimate (comparing our four sampled settlements): Mean pairwise FST = 0.003 (95% CI: 0.001–0.007); interpretation: Minimal genetic structure across Tawahka settlements, consistent with high gene flow and cultural cohesion.
• Between-population estimate (comparing Tawahka with Lenca and Honduran mestizo): (i) FST (Tawahka vs. Lenca) = 0.012 (95% CI: 0.008–0.018), (ii) FST (Tawahka vs. Honduran mestizo) = 0.025 (95% CI: 0.019–0.033) and (iii) FST (Tawahka vs. NIST Caucasian) = 0.095 (95% CI: 0.087–0.104).

Based on these estimates and simulation studies assessing the impact on RMP calculations, we recommend population-specific θ values shown in

Table 1. Recommended Theta values for forensic calculations in Honduras.

Scenario	Recommended θ	Justification
Within Tawahka (suspect and reference both confirmed Tawahka)	θ = 0.01	Conservative given within-population FST = 0.003; accounts for potential cryptic relatedness
Within Lenca	θ = 0.01	Similar reasoning
Within Honduran mestizo	θ = 0.01	Existing standard; appropriate given moderate admixture and urbanization
Tawahka vs. mestizo (uncertain ethnicity)	θ = 0.03	Reflects elevated FST = 0.025; conservative approach for cross-population comparisons
Tawahka vs. European/Asian (international cases)	θ = 0.05–0.10	Reflects substantial FST; highly conservative
Unknown ethnicity (no information)	θ = 0.03	Intermediate value balancing conservatism with informativeness

.

To illustrate the practical effect, we calculated RMP for a representative Tawahka profile using different θ values:

• θ = 0.00 (no correction): RMP = 1 in 8.7 × 10¹⁵;
• θ = 0.01 (current standard): RMP = 1 in 5.2 × 10¹⁵ (1.7 × higher);
• θ = 0.03 (recommended for cross-population): RMP = 1 in 1.8 × 10¹⁵ (4.8 × higher);
• θ = 0.05: RMP = 1 in 9.1 × 10¹⁴ (9.6 × higher).

While these adjustments modestly reduce the strength of evidence, all values remain exceptionally strong. The key principle is that θ correction ensures conservative, defensible calculations that account for potential relatedness or substructure. We propose that the Forensic Medicine Directorate of Honduras adopt a tiered theta framework based on ethnicity and population affinity:

• Tier 1 (within-population): θ = 0.01 for same ethnic group;
• Tier 2 (between indigenous groups): θ = 0.02–0.03;
• Tier 3 (indigenous vs. mestizo): θ = 0.03;
• Tier 4 (unknown or international): θ = 0.03–0.05.

This approach balances statistical conservatism with informativeness, ensuring that likelihood ratios remain meaningful while adequately accounting for population structure. Implementation would require, (i) Standardized ethnicity documentation in case files; (ii) Training for forensic analysts on theta selection and, (iii) Clear guidelines for expert testimony explaining theta corrections to courts.

Future studies with larger sample sizes and genome-wide markers will refine these estimates and enable more precise FST calculations for Honduras’s diverse populations [58]. The product rule for calculating combined match probabilities assumes that genotypes at different loci are independent [3]. This assumption is generally robust for unlinked STR markers in equilibrium populations. However, departures from independence can arise from:

• Physical linkage: STRs on the same chromosome within ~50 cM may show residual linkage disequilibrium. Our 23 markers span 14 autosomes with pairwise distances > 50 cM, ensuring independence [12,18].
• Population admixture: Recent admixture creates genome-wide correlations that decay slowly over generations [59]. The Tawahka’s relative isolation suggests limited recent admixture, though historical mixing cannot be excluded.
• Selection or epistasis: If STR-linked regions are under selection, correlations could arise. However, STRs are generally neutral [1,2].

To assess multi-locus dependencies empirically, we calculated pairwise linkage disequilibrium (LD) for all 253 locus pairs using the composite measure D′. Only 4/253 pairs (1.6%) showed |D′| > 0.2, and none exceeded |D′| = 0.35, indicating minimal LD. These results support the independence assumption underlying the product rule.

Nevertheless, we emphasize that forensic statistics should be interpreted with appropriate humility. Reporting RMP = 1 in 10²⁰ implies precision that exceeds our actual knowledge of population structure, locus independence, and database accuracy. We advocate combined reporting statistics as “>1 in 10^X” where X reflects the magnitude of evidence (e.g., >1 in 10¹⁴) rather than spurious precision [48]. This approach aligns with recommendations from the National Research Council [3] and the DNA Commission of the ISFG [60].

4.5. Comparison with Other Populations

Although formal inter-population analyses lie beyond the scope of this data article, preliminary patterns are consistent with those reported for other Central American Indigenous groups [53,55]. Several allele frequencies diverge from values commonly observed in European and Asian reference populations, underscoring the need for population-specific databases in forensic inference [61,62]. The detection of rare alleles may further reflect the relative genetic isolation of the Tawahka community. Building on this baseline, future work should quantify genetic relationships with other Honduran Indigenous groups, characterize admixture with neighboring populations, and investigate the evolutionary and demographic history of the Tawahka in the context of broader Central American diversity.

To contextualize the genetic diversity of the Tawahka, we compared key forensic parameters with previously published Central American datasets (

Table 2. Comparative forensic parameters across Central American populations.

Population	N	Mean He	Mean PD	Mean Alleles/Locus	Reference
Tawahka (Honduras)	100	0.7385	0.8771	8.61	Present study
Lenca (Honduras)	100	0.7425	0.8815	8.91	[9]
Maya (Guatemala)	127	0.7104	0.8534	7.83	[53]
Ladino (Guatemala)	115	0.7656	0.8945	9.45	[55]
Mestizo (El Salvador)	108	0.7512	0.8823	8.95	[63]

).

The Tawahka exhibited genetic diversity parameters (mean He = 0.7385) nearly identical to the geographically proximate Lenca (He = 0.7425), suggesting shared demographic history or similar effective population sizes. Both Honduran indigenous groups showed slightly lower diversity than admixed Guatemalan Ladino populations (He = 0.7656), consistent with expectations for smaller, more isolated communities [10]. The slightly reduced allelic richness in Tawahka (8.61) compared to Lenca (8.91) may reflect sampling effects or genuine demographic differences requiring further investigation.

Specific allele frequency patterns support this regional affinity. For example, at the D7S820 locus, allele 10 frequencies were 0.31 (Tawahka), 0.29 (Lenca), and 0.25 (Maya), versus 0.18 in European populations [61,62]. Similarly, at D22S1045, allele 16 exhibited frequencies of 0.60 (Tawahka), 0.58 (Lenca), and 0.52 (Maya), compared to 0.35–0.40 in European and Asian reference datasets [62].

The detection of population-specific rare alleles (e.g., D21S11 variants 27.3 and 30.3; SE33 alleles 8 and 32.2) further underscores the necessity of population-specific databases. Forensic calculations using generic ‘Hispanic’ or European reference data could yield biased likelihood ratios for individuals of Tawahka ancestry [3,4].

Future work should include formal FST analysis across multiple Honduran and Central American populations, as well as admixture analysis using genome-wide markers to quantify European, Indigenous American, and African ancestry components [64]. Such analyses would clarify whether the Tawahka represent a genetically distinct unit or form part of a broader eastern Honduran indigenous cluster.

4.6. Practical Applications

This dataset has broad practical utility across forensic, population, anthropological, and medical genetics. In forensic applications, it provides a population-specific reference panel for likelihood-ratio calculations, supports paternity and broader kinship assessments within the Tawahka community, and informs missing-persons and disaster victim identification workflows. From a population-genetic standpoint, the allele frequencies enable analyses of Indigenous Central American diversity, facilitate tests of migration and admixture, and contribute to reconstructions of evolutionary and demographic history. Anthropologically, the data offers genetic context for evaluating the origins, affinities, and potential correspondence between cultural or linguistic groupings in Mesoamerica. Finally, as a foundational medical resource, the dataset can guide hypothesis-driven studies on genetic disease predisposition and inform future pharmacogenetic approaches tailored to Indigenous populations. The following subsections detail the forensic impact and implementation plans.

4.6.1. Forensic Impact of Population-Specific Databases

The use of population-specific allele frequencies directly impacts the accuracy and legal defensibility of forensic likelihood ratio (LR) calculations [3,4]. To illustrate this impact for the Tawahka population, we compared match probabilities calculated using three reference databases: (1) Tawahka-specific frequencies (this study), (2) general Honduran mestizo frequencies [16], and (3) a generic ‘Hispanic’ database commonly used in forensic casework [17].

For a simulated profile typical of the Tawahka population (with genotypes at high-frequency alleles such as D2S441 10/10, D22S1045 16/16, and TPOX 8/11), the random match probability (RMP) was:

• Tawahka database: 1 in 8.2 × 10¹⁵;
• Honduran mestizo database: 1 in 4.1 × 10¹⁶ (5 × difference);
• Generic Hispanic database: 1 in 1.3 × 10¹⁷ (16 × difference).

While all three databases yield exceptionally strong evidence of identity, the differences are forensically meaningful. Using non-specific databases for Tawahka individuals systematically overestimates the strength of evidence, potentially misleading triers of fact. Conversely, for admixed Honduran individuals analyzed with the Tawahka database, the evidence strength would be underestimated, potentially benefiting defendants inappropriately.

More critically, in paternity cases, the use of inappropriate reference data can alter exclusion probabilities and paternity indices significantly. For a standard trio (mother-child-alleged father), the paternity index (PI) calculated using the Tawahka database differed by 15–40% compared to generic Hispanic frequencies at specific loci (D7S820, D22S1045, D2S441) where allele frequency differences exceeded 0.20.

These findings underscore the legal and scientific imperative for population-specific databases, particularly for indigenous communities with distinct allele frequency distributions. Forensic laboratories serving regions with indigenous populations should maintain curated databases stratified by ethnicity to ensure accurate, ethically sound, and legally defensible analyses [65,66].

4.6.2. Impact Analysis Using Actual Case Profiles

To quantify the practical impact of the Tawahka-specific database, we retrospectively analyzed five anonymized forensic case profiles from the Gracias a Dios region (2020–2023) where individuals were of suspected Tawahka ancestry. For each profile, we calculated random match probabilities (RMP) using three databases: (1) Tawahka (this study), (2) Honduran mestizo [16], and (3) generic Hispanic [17]. Theta correction (θ = 0.01) was applied consistently [18]. All this can be found in

Table 3. Comparative match probabilities for five case profiles.

Case	RMP (Tawahka)	RMP (Mestizo)	RMP (Hispanic)	Fold Difference *
A	1 in 4.2 × 1014	1 in 1.8 × 1015	1 in 6.5 × 1015	4.3×–15.5×
B	1 in 8.7 × 1015	1 in 3.1 × 1016	1 in 9.8 × 1016	3.6×–11.3×
C	1 in 1.1 × 1013	1 in 3.9 × 1013	1 in 1.4 × 1014	3.5×–12.7×
D	1 in 2.9 × 1016	1 in 1.2 × 1017	1 in 4.1 × 1017	4.1×–14.1×
E	1 in 6.5 × 1014	1 in 2.7 × 1015	1 in 8.2 × 1015	4.2×–12.6×

* Fold difference = (RMP mestizo or Hispanic)/(RMP Tawahka).

.

With Table 2, main key findings are:

• The Tawahka database consistently yielded higher (less incriminating) RMP values compared to non-specific databases, with fold differences ranging from 3.5× to 15.5×.
• This pattern reflects the presence of higher-frequency alleles in the Tawahka population at specific loci (e.g., D2S441 allele 10: 0.66 vs. 0.42 in mestizos; D22S1045 allele 16: 0.60 vs. 0.38 in mestizos).
• While all RMP values remain exceptionally small (strong evidence of identity), the differences have legal significance. For example, in Case C, the difference between 1 in 1.1 × 10¹³ (Tawahka) and 1 in 1.4 × 10¹⁴ (Hispanic) represents a 12.7-fold overstatement of evidence strength if the wrong database is used.
• In paternity testing, similar effects were observed. For three paternity cases involving Tawahka families, the combined paternity index (CPI) varied by 18–35% depending on the database used, with the Tawahka database yielding lower (more conservative) CPI values.

The results underscore that the choice of reference database is not a minor technical consideration but a critical factor influencing the evidentiary weight of genetic findings in forensic contexts. The use of non-specific or inappropriate databases for individuals of indigenous origin can systematically inflate the strength of genetic evidence, thereby increasing the risk of wrongful convictions or erroneous paternity determinations. Conversely, the lack of population-specific allele frequency data compels forensic practitioners to either rely on unsuitable reference populations or refrain from performing statistical analyses altogether, both of which compromise the fairness and reliability of judicial processes. In this context, the establishment of the Tawahka reference database fulfills two essential objectives: first, it enables accurate and ethically sound forensic evaluations for this community; second, it provides a methodological framework for the systematic genetic characterization of other underrepresented indigenous populations in Honduras.

4.6.3. Additional Applications

Beyond forensic casework, this dataset enables:

• Population genetics: The allele frequencies facilitate analyses of Indigenous Central American diversity, tests of migration and admixture, and reconstructions of evolutionary and demographic history.
• Anthropology: The data offers genetic context for evaluating the origins, affinities, and potential correspondence between cultural or linguistic groupings in Mesoamerica.
• Medical genetics: As a foundational resource, the dataset can guide hypothesis-driven studies on genetic disease predisposition and inform future pharmacogenetic approaches tailored to Indigenous populations.

4.7. Limitations and Considerations

This dataset should be interpreted considering several limitations. First, although a sample of 100 individuals is adequate for many forensic applications, larger cohorts would improve precision for population-genetic parameters and rare-allele estimates. Second, although samples were collected from four settlements within the municipality of Wampusirpi (Krausirpi, Yapuwas, Brisas de Pisijire, and Krautura), this geographic scope is restricted to a single municipality within the eastern Tawahka range and may not capture the full spatial heterogeneity of the Tawahka population distributed across the departments of Olancho and Gracias a Dios. Third, the data provides a temporal snapshot and therefore cannot resolve historical dynamics or ongoing demographic change. Fourth, formal population structure analysis (e.g., FST, PCoA) was not conducted due to the unavailability of raw genotype data from comparison populations; future collaborative efforts should establish a Central American STR database with shared individual-level profiles to enable robust inter-population analyses. Finally, some degree of admixture with neighboring groups is plausible as in many contemporary Indigenous populations, despite deliberate enrollment of participants with documented Tawahka ancestry over at least three generations; residual admixture could subtly influence allele frequencies and should be explored in future, geographically broader studies.

4.8. Statistical Considerations and Multiple Testing

The application of Bonferroni correction across 23 loci warrants careful interpretation. This conservative approach prioritizes specificity over sensitivity, reducing false-positive detection of HWE deviations at the cost of potentially missing genuine but subtle departures. In our Tawahka dataset, no locus showed significant deviation after correction (all p > 0.0870), suggesting either true equilibrium or insufficient power to detect minor effects.

For forensic applications, this conservative stance is appropriate, we prioritize avoiding the inclusion of potentially problematic loci in likelihood ratio calculations. However, researchers using this dataset for population genetic inference should be aware that mild violations of HWE or weak evidence of substructure may remain undetected. Future studies with expanded sample sizes (N > 200) could employ less stringent corrections or model-based approaches to better characterize population structure [37,38].

4.9. Implementation and Future Directions

The Tawahka allele frequency database will be formally incorporated into the operational database of the Forensic Medicine Directorate, Public Ministry of Honduras, effective January 2026. Forensic analysts will have access to Tawahka-specific frequencies through the laboratory information management system (LIMS). When circumstances or self-reported ethnicity indicate Tawahka ancestry, analysts will preferentially employ this database for statistical calculations.

To ensure appropriate implementation, we have developed a training module for forensic personnel covering: (1) appropriate database selection based on ethnic background; (2) statistical methods for incorporating population-specific frequencies; (3) ethical considerations in genetic ancestry inference; and (4) expert testimony guidelines for explaining database selection in court.

Building upon the methodological framework established in the present study, the Public Ministry and the National Autonomous University of Honduras (UNAH) have launched a series of parallel population genetics projects (2025–2028) targeting five additional indigenous groups: Miskito (target N = 100, eastern Honduras), Pech (target N = 100, northeastern Honduras), Tolupán (target N = 100, central Honduras), Chortí (target N = 100, western Honduras), and Garifuna (target N = 100, southern Honduras), along with an expanded mestizo dataset (target N = 200, northern and central regions). Collectively, these efforts aim to establish comprehensive STR frequency databases for all major Honduran indigenous populations by 2028, thereby promoting equitable access to accurate and culturally sensitive forensic genetic services regardless of ethnic background.

In parallel, a regional initiative is underway to create a Central American Indigenous STR Database Network through collaboration with forensic laboratories in Guatemala, El Salvador, Nicaragua, and Belize. This consortium will facilitate cross-border forensic investigations involving mobile indigenous populations, enable robust analyses of regional population structure through standardized analytical protocols, harmonize statistical methodologies across jurisdictions, and strengthen regional capacity through technical exchange and collaborative training among forensic scientists.

The Tawahka allele frequency data generated in this study have been submitted to international repositories, including STRBase (NIST, Gaithersburg, MD, USA) [49] and the European Network of Forensic Science Institutes (ENFSI) DNA Working Group database [67], ensuring global accessibility and transparency. Additionally, a submission to the EMPOP database is being prepared to support forthcoming mitochondrial DNA research initiatives [68].

Recognizing that allele frequencies may vary over time due to admixture, genetic drift, or demographic shifts, a long-term quality assurance plan has been implemented. This plan includes comprehensive database reviews every 10–15 years, complemented by interim targeted sampling when significant demographic changes are detected, to maintain the accuracy and relevance of population data.

Finally, in partnership with the Tawahka Indigenous Federation (FITH), a community outreach and education program is being developed to enhance local understanding of forensic genetics. Educational materials will explain the purpose and applications of genetic databases in forensic science, outline privacy and data protection measures, describe the benefits of population-specific databases, and emphasize the community’s ongoing role in research oversight. These materials will be disseminated through participatory workshops, bilingual radio broadcasts in Tawahka and Spanish, and accessible written formats tailored to diverse literacy levels, ensuring meaningful engagement and ethical stewardship of genetic research.

5. Conclusions

This study delivers the first comprehensive genetic characterization of the Honduran Tawahka based on 23 autosomal STR loci. The panel reveals high within-population diversity, with all markers exhibiting substantial polymorphism and no significant departures from Hardy–Weinberg equilibrium after multiple-testing correction, consistent with genetic equilibrium. Forensic performance was exceptional: the combined power of discrimination exceeded 99.999% and the combined chance of exclusion likewise surpassed 99.999%, underscoring the suitability of the PowerPlex Fusion 6C System for both individualization and kinship analysis in this population. These results provide a much-needed population-specific reference for forensic casework, while also furnishing baseline data for population-genetic and anthropological research in Central American Indigenous groups. All genetic profiles were stored in encrypted form following CODIS and ESS standards to safeguard identifiability. To our knowledge, this is the first report of autosomal STR markers in a Tawahka cohort, addressing a critical gap in the regional genetic literature.