Genetic Identification of Human Remains in Mexico
An Update on DNA Extraction and STR Typing on Degraded Bones
Valentina Leonie Birne1, Franziska Holz1, Marcel A. Verhoff1, Christoph G. Birngruber1, Richard Zehner1 1 Institute of Legal Medicine, University Hospital Frankfurt, Goethe University, Frankfurt am Main, Germany
Share this article
INTRODUCTION
1. Forensic crisis in Mexico
As described in our previous articles (“Genetic identification of human remains in Mexico: Increasing DNA extraction efficiency from degraded tissues using the Promega Maxwell® RSC 48 instrument.” [1] and “Genetic identification of human remains in Mexico: Increasing STR allele calling from degraded tissues using the Promega 8-dye PowerPlex® 35GY System.” [2], published in the Spring and Autumn 2024 issues of Profiles in DNA), authorities declared a forensic crisis in Mexico. Since the beginning of the so-called ‘war on drugs’ in 2006, violence in Mexico has increased, mainly caused by organised criminal groups and the fight against them [3,4]. To put in figures: more than 125,000 people are currently missing in Mexico, over 72,000 bodies are unidentified, and countless illegal graves are unprocessed [5,6]. It is estimated that more than 150,000 people have been killed in Mexico between 2006 and 2018, and most of them remain unidentified [7]. Amy Reed-Sandoval calculated in 2024 that it would take forensic experts up to 120 years to identify all the human remains of the disappeared [8]. Based on these figures, the authorities speak of a forensic and humanitarian crisis related to the identification of unknown deceased persons.
2. Identification in Mexico
Identification of unknown deceased persons or body parts in Mexico is an interdisciplinary cooperation. At the site of the deceased, for example, medical doctors run the autopsy, criminologists take fingerprints, anthropologists estimate the body height, odontologists document the dental records, and biologists obtain samples for DNA analysis. At the site of the living, families report their missing relatives to the authorities. A questionnaire is then filled out about the person missing, with descriptions of medical conditions, appearance, and pictures. In addition, DNA samples from the family members are taken for a later comparison if an unidentified body or body parts are found that matches the description of the family [9].
The postmortem and ante mortem data described above are collected and compared by the authorities. However, DNA analysis is often the only approach left for sufficient identification, as only bones or parts of bodies are left of the deceased and no dental status, fingerprints, or secondary identifiers (personal belongings or physical appearance) can be collected.
3. German-Mexican international collaboration
3.1 Project “CoCiMex” (DAAD)
The “CoCiMex” project was founded in 2021 as a German-Mexican University collaboration to support the Mexican Forensic Medical Services in the identification process of unknown deceased individuals in Mexico. The Institute of Legal Medicine in Frankfurt, Germany, was among others collaborating on this project. The research project was supported by DAAD with funds from the German Cooperation for Sustainable Development GmbH on behalf of the German Federal Foreign Office (DAAD Project ID: 57594060).
3.2 Project “IDH” (UNFPA)
As follow-up project after the end of the CoCiMex cooperation, UNFPA and Goethe University Frankfurt am Main signed a cooperation agreement at the end of June 2023 to support the Mexican government in identifying unknown bodies in Mexico. The current project is called “IDH”: Identificación Humana en México.
4. DNA analysis from bones
As mentioned above, bones are often the only tissue left of human remains in Mexico, due to criminal activity and/or natural decomposition. DNA in bones is well preserved due to the robust mineral matrix that acts as physical, chemical and microbial barrier that protects DNA from external influences as well as chemical and microbial degradation [10,11]. Unfortunately, the hard and dense bone structure makes it difficult to extract DNA from bones. Thus, processing bones in the laboratories is very time consuming and costly . First, the bones have to be physically or chemically decontaminated, then the bones need to be pulverised. After the initial preparation, various manual or automated DNA extraction protocols can be applied, which may take several days of incubation and hours of labour.
As the success of STR allele typing depends on sufficient DNA quality and quantity, it is crucial to find the most suitable DNA extraction protocol for your laboratory. To improve the success of STR typing of degraded bones in our laboratory, we compared four different DNA extraction protocols and analysed the extracted DNA using the standard 5-dye STR kit and the new 8-dye STR kit from Promega.
MATERIALS AND METHODS
1. DNA extraction of bones
Different types of bones such as clavicle, humerus, vertebrae, femur, metacarpals and metatarsals were collected from various parts of the body in different states of decomposition during an autopsy in Mexico. The body parts were exhumed from a mass grave whose burial time, storage conditions and pretreatment are unknown. To increase the efficiency of DNA extraction from these post mortem-altered tissues, DNA from 15 of these bones were extracted with four different DNA extraction methods: the classical organic “Phenol-Chloroform” extraction, two extraction methods “Dabney” and “Loreille” originally from the anthropology designed for ancient skeletal remains, and one adapted half-automated extraction method using the Maxwell® RSC 48 instrument from Promega (“Maxwell®”). All 15 bones were pulverised once and then divided for the four different methods.
1.1 Loreille
The Loreille extraction [12], originally developed for anthropological research, is based on complete demineralised and dissolved bone powder. The principle of completely dissolving the bone material should ensure that the whole enclosed DNA can be released from the bone matrix. After complete demineralisation and cell lysis, the lysate is purified using silica-based columns. The Loreille protocol was performed as described by Emery et al. 2020 [13] with 200 mg bone powder.
1.2 Maxwell® Instrument
An adapted protocol with the Maxwell® RSC 48 instrument (Promega, Madison, USA) using the Maxwell® FSC DNA IQ™ Casework kit combined with the Bone DNA Extraction Kit (Promega, Madison, USA) was used for the Maxwell® extraction. Here, the use of two 100 mg doses of bone powder, i.e. a total of 200 mg bone powder, and the pooling of bone samples lead to an increase in DNA extraction. For further information, we recommend our previous article published in the Spring issue 2024 of Profiles in DNA [1] and the initial publication from Duijs and Sijen et al. 2020 [14]. The half-automated DNA extraction protocol is based on magnetic beads that selectively bind DNA and purifies the bound DNA during different steps (see Maxwell® FSC DNA IQ™ Casework Kit Technical Manual ).
1.3 Dabney
The original Dabney [15] protocol was also developed for anthropological research and is similar to the Loreille protocol. The method is also based on two steps: decalcification and cell lyses followed by DNA purification using silica-matrices. In this study, we used an adapted Dabney based protocol which was kindly provided to us by colleagues from the Institute of Legal Medicine in Kiel, Germany. First, 100 mg bone powder was decalcified and lysed overnight, using EDTA and proteinase K. Then, the lysate was purified using the Pure Viral Nucleic Acid Large Volume Kit (Roche) which is based on glass fibre membrane columns.
1.4 Phenol-Chloroform
Phenol-Chloroform is a classic organic DNA extraction method that has long been used in forensics for various tissue types. The extraction is based on the different solubility of different molecules in aqueous and organic phases to separate nucleic acids from proteins and lipids [16]. Nucleic acids remain in the aqueous phase, while proteins and lipids are removed. 500 mg bone powder was used for this Phenol-Chloroform protocol.
2. DNA quantification
DNA quantification was performed with the PowerQuant® System qPCR assay (Promega, Madison, USA). To determine the DNA degradation index (DI), the ratio of the calculated concentration values of the amplicons (short amplicon/long amplicon) is calculated (PowerQuant® System Technical Manual). The higher the degradation value, the higher is the degradation of the DNA of interest. From a degradation index of 6, effects on the STR profile quality in the long fragment range can be observed on regular casework samples (in-house validation). The qPCR assay was performed on a 7500 Real-Time PCR System (Applied Biosystems, Waltham, USA) according to the manufacturer’s guidelines with half of the recommended reaction volume.
3. STR allele calling 5-dye vs. 8-dye strategy
To improve STR allele calling, we compared a standard 5-dye STR kit with a new 8-dye STR kit. The following assays were performed on a Spectrum Compact System (Promega, Madison, USA) according to the manufacturer’s guidelines with half of the recommended reaction volume. The generated data was analysed using the GeneMarker®HID Software for Spectrum CE Systems (Promega, Madison, USA).
3.1 5-dye strategy
Routinely, we use the PowerPlex® Fusion System (Promega, Madison, USA) for typing STR alleles from Mexican samples, as it meets the CODIS Core Loci and the European Standard Set. In a 24-Plex it includes 22 autosomal STR markers, amelogenin, and one Y-STR for sex identification; see Figure 1.
Figure 1: Overview of the 5-dye PowerPlex® Fusion System.
3.2 8-dye strategy
To improve the STR allele calling from forensic challenging samples, the 8-dye PowerPlex® 35GY System (Promega, Madison, USA) was compared against the routinely used PowerPlex® Fusion 5-dye kit. The PowerPlex® 35GY System is designed to detect the 20 CODIS core loci plus Amelogenin, Penta D, Penta E, SE33 and 11 Y-STR loci, as well as quality indicators by using an 8-dye short amplicon strategy (PowerPlex® 35GY System Technical Manual); see Figure 2. Two additional dye channels allow 15 autosomal loci to have less than 250 bp, which can be advantageous for degraded forensic samples.
Figure 2: Overview of the 8-dye PowerPlex® 35GY System
RESULTS
1. DNA quantity and quality
The DNA yield of all 15 samples extracted using the four extraction methods presented (Loreille, Maxwell®, Dabney and Phenol-Chloroform) is shown below in Figure 3. As the amount of bone powder and the elution volume were different for each method, the normalised DNA yield (pg DNA per mg bone powder used) was used for comparison.
In this study, samples extracted with the Loreille method achieved the highest DNA yield with an average DNA yield of 22.4 pg/mg, followed by Dabney, Maxwell® and, with the lowest yield, Phenol-Chloroform with 2.7 pg/mg DNA, as shown in Figure 3A. The DNA DI for the samples extracted with Loreille, Maxwell® and Dabney was similar (see Figure 3B), in contrast to the same samples extracted via Phenol-Chloroform. Here the DI is higher than for the samples extracted with the other methods.
Figure 3: DNA yield and DI of 15 bone sample extracted with four different DNA extraction methods. The different extraction methods are illustrated in different colours: Loreille (pink), Maxwell (orange), Dabney (blue) and Phenol-chloroform (green). Figure 3A depicts the DNA yield (pg DNA per mg bone powder used) with calculated average values “Ø” and Figure 3B depicts the DNA DI with calculated average values “Ø”. The bar shows the calculated mean.
2. 5-dye vs 8-dye strategy
Since we not only wanted to test which DNA extraction method results in the highest DNA quantity and quality, but also how the STR profile quality is affected, we analysed all extracted samples with the 5-dye strategy using the PowerPlex® Fusion and the 8-dye PowerPlex® 35GY, Systems shown in Figure 4.
Figure 4: Number of called STR loci using 5-dye and 8-dye strategies of 15 bones extracted with four different DNA extraction methods. Amount and calculated average “Ø” of STR loci analysed with the PowerPlex® Fusion (5-dye system) and PowerPlex® 35GY (8-dye system) are shown. The number of autosomal STR loci analysed with the 5-dye system is marked in grey, the number of autosomal STR loci analysed with the 8-dye system is shown in colour, and the number of analysed Y-STR loci is shown in the corresponding lighter colour: (A) Loreille, pink/ light pink; (B) Maxwell®, orange/ light orange; (C) Dabney, blue/ light blue and (D) Phenol-Chloroform, green/ light green. The bar shows the calculated mean.
First, most STR markers could be called from samples extracted with the Loreille method with an average of 20 autosomal STR marker with the 5-dye strategy and an average of 22 autosomal loci with the 8-dye strategy, followed by the Maxwell® method and the Dabney method. Bones extracted using the Phenol-Chloroform method achieved the lowest number of STR markers called for both dye strategies.
Second, the overall performance of the 8-dye kit was better when compared to the 5-dye kit for all bone samples and extraction methods tested. However, both a complete 5-dye PowerPlex® Fusion STR profile and a complete 8-dye PowerPlex® 35GY System STR profile could be generated with the same bone with a minimum of 175.5 pg DNA input (24.3 pg/µl, DI 2.78). The 8-dye PowerPlex® 35GY System STR profile of this sample is shown in Figure 5.
Figure 5: An example of a complete PowerPlex® 35GY System profile of a bone with the lowest DNA input of 175.5 pg. The bone was extracted using the Dabney method and analysed with the PowerPlex® 35GY System and the GeneMarker®HID Software.
CONCLUSION
1. Which method is the best for DNA extraction?
In our initial study, to find the best-suited DNA extraction method for degraded bones we compared four different methods: two manual extraction protocols (Loreille and Dabney designed for DNA extraction of ancient skeletal remains), the classical Phenol-Chloroform protocol, and one half-automated protocol using the Maxwell® instrument. The method with the lowest DNA yield, highest DNA degradation index and lowest number of called STR markers was Phenol-Chloroform. The high degradation indices can be explained, for example, by residues of the reagents in the extract, which also have an influence on the STR profile. It is a very favourable method, but it is outdated in many laboratories due to the needed hazardous reagents. Based on our available findings, we can therefore remove Phenol-Chloroform DNA extraction for bones from our portfolio. The highest DNA yield could be achieved using the two anthropology protocols Loreille and Dabney, followed by Maxwell®. Samples extracted with the Loreille protocol also generated the highest number of STR loci, followed by samples extracted with the Maxwell® protocol. On the one hand, Dabney was not the second best, as the Maxwell® extracts are very clean and free from reagent residues which favours the STR amplification. On the other hand, the sample with the lowest amount of DNA and yet a complete STR profile was extracted using the Dabney method. The profile of the same sample extracted with Loreille and Maxwell® was also complete, but had a higher DNA yield. The results indicate that there may not be one perfect extraction method that is suitable for every bone. For difficult bone samples, more than one method probably needs to be tested to find the most suitable one for the sample. This was likewise discussed by Edson 2019, who examined the connection of chemical residues and DNA recovery in skeletal remains [17]. For example, for one bone the DNA yield achieved with the Dabney protocol was 34.6 pg/mg, with the Maxwell® protocol a DNA yield of 20.63 pg/mg was achieved and with the Loreille extraction a DNA yield of 10.08 could be extracted (data not shown), although the Loreille extraction was usually best for the remaining samples. This could be due to the fact that the sample was treated with chemicals that clogged the filters of the Amicon Ultracel membrane tubes used in the Loreille method.
As we could not find a definitive answer to the question of which extraction method is the best, we still have a lot of work to do. For now, however, we are using the Maxwell® extraction protocol when processing bones, as it is the simplest, quickest and cleanest method tested. For difficult samples, we still need to make further adjustments and validations to develop an optimal manual protocol for highly degraded skeletal remains based on the original Dabney [15] and Loreille [12] extraction methods.
2. Can the 8-dye strategy improve STR allele typing?
To come back to the question posed at the beginning: Can the PowerPlex® 35GY improve the analysis of STR alleles in degraded bone? To answer the last question, we summarise the results of this study, which underline our previous statement in the Fall 2024 issue [2]. Overall, the general performance of the 8-dye kit was better compared to the 5-dye kit for all bone samples and extraction methods tested. This is assumed because most of the markers of the 8-dye system have a small base pair size under 250 bp, which is a great advantage especially in this degraded bone material. In addition, the heterozygous alleles were balanced and the 8-dye kit was very sensitive despite 35 markers in a multiplex PCR. With a minimum of 175.5 pg DNA (24.3 pg/µl) and a DI of 2.78 a complete profile with 35 allele loci could be generated. The same sensitivity was achieved for the 5-dye PowerPlex® Fusion System with 24 allelic loci in our sample set. The results show that the 8-dye strategy has the same sensitivity despite the three additional colour channels and 11 additional allelic markers. In the best case, you can therefore obtain more markers with one analysis: 10 Y-STR markers and SE33.
Furthermore, we see the advantage of a kit containing both autosomal and Y-STR markers in 8-colour channels for forensic kinship and DNA mixture analyses. In addition, the 8-dye short amplicon strategy can be helpful when working with degraded and/or inhibited DNA. Therefore, especially in difficult cases, obtaining markers with short amplicon lengths can be a great advantage.
3. Project update – how far have we come?
As the main aim of the projects and the associated cooperation with our German and Mexican partners is the identification of unknown deceased and to return them to their relatives, we can report some positive identifications.
Again in figures: In 2022, 431 body parts were exhumed; using classic forensic STR analysis, we were able to reconstruct and identify 38 deceased persons with our colleagues from Mexico. The identified deceased were handed over to their families.
In collaboration with the Institutes of Forensic Medicine in Münster, Germany, and Giessen, Germany, we have so far been able to analyse DNA samples from various Mexican states, including Jalisco, San Luis Potosí, Sinaloa, Tamaulipas and Zacatecas. A total of 2,164 ante mortem samples from family members who had reported their loved ones missing and 837 post mortem samples from deceased persons in Mexico have been analysed. These profiles were handed over to the Mexican authorities and serve as support for the local identification process.
In addition to the samples that were processed to support our colleagues in Mexico, we were able to achieve research results that are adapted to the situation and possibilities on site. One example and hands-on study is the potential of tattoos in the identification process [18,19].
A tool has been developed to classify tattoo motives in Mexico as tool to identify unknown bodies [18] that demonstrates a highly efficient and cost-effective identification strategy. Another study focussed on the use of tendons as alternative DNA material to reduce the cost and time required for DNA analysis. In addition, table salt was tested as preservative for tendons for subsequent DNA analysis when no cooling on site is available [20].
REFERENCES
[1] Birne V.L., Holz F., Verhoff M.A., Birngruber C.G., Zehner R. Genetic Identification of Human Remains in Mexico Increasing DNA Extraction Efficiency from Degraded Tissues Using the Promega Maxwell® RSC 48 Instrument. Profiles in DNA 2024;Volume 09. [2] Birne V.L., Holz F., Verhoff M.A., Birngruber C.G., Zehner R. Genetic Identification of Human Remains in Mexico Increasing STR allele calling from degraded tissues using the Promega 8-dye PowerPlex® 35GY System. 2024;Volume 10. [3] Del Pilar Fuerte Celis M, Lujan EP, Ponce RC. Organized crime, violence, and territorial dispute in Mexico (2007–2011). Trends Organ Crim 2019;22:188–209. https://doi.org/10.1007/s12117-018-9341-z. [4] Atuesta LH, Siordia OS, Lajous AM. The “War on Drugs” in Mexico: (Official) Database of Events between December 2006 and November 2011. Journal of Conflict Resolution 2019;63:1765–89. https://doi.org/10.1177/0022002718817093. [5] San Juan Flores P, Guillén B. México, el país que desaparece: sin rastro de 125.000 personas 2025. [6] Tzuc E, Sánchez M. Cierra sexenio con más de 72,100 cuerpos sin identificar. A donde van los desaparecidos 2024. https://adondevanlosdesaparecidos.org/2024/09/24/cierra-sexenio-con-mas-de-72100-cuerpos-sin-identificar/ (accessed June 30, 2025). [7] International comission of missing persons. According to the National Registry of Missing Persons, as of 21 September 2023, 111,521 persons were reported as disappeared in Mexico as a result of crime. n.d. https://icmp.int/what-we-do/geographic-programs/mexico/ (accessed July 2, 2025). [8] Amy Reed-Sandoval. The Struggle to Identify All the Dead Bodies in Mexico. The New Yorker 2025. [9] Flores L. Familiares de personas desaparecidas exigen Banco Nacional de Datos Forenses 2022. [10] Latham KE, Miller JJ. DNA recovery and analysis from skeletal material in modern forensic contexts. Forensic Sci Res 2019;4:51–9. https://doi.org/10.1080/20961790.2018.1515594. [11] Sorg MH, Haglund WD, editors. Forensic Taphonomy. CRC Press; 1996. https://doi.org/10.1201/9781439821923. [12] Loreille OM, Diegoli TM, Irwin JA, Coble MD, Parsons TJ. High efficiency DNA extraction from bone by total demineralization. Forensic Sci Int Genet 2007;1:191–5. https://doi.org/10.1016/j.fsigen.2007.02.006. [13] Emery MV, Bolhofner K, Winingear S, Oldt R, Montes M, Kanthaswamy S, et al. Reconstructing full and partial STR profiles from severely burned human remains using comparative ancient and forensic DNA extraction techniques. Forensic Sci Int Genet 2020;46:102272. https://doi.org/10.1016/j.fsigen.2020.102272. [14] Duijs FE, Sijen T. A rapid and efficient method for DNA extraction from bone powder. Forensic Sci Int Rep 2020;2:100099. https://doi.org/10.1016/j.fsir.2020.100099. [15] Dabney J, Meyer M. Extraction of Highly Degraded DNA from Ancient Bones and Teeth. In: Shapiro B, Barlow A, Heintzman PD, Hofreiter M, Paijmans JLA, Soares AER, editors. Ancient DNA, vol. 1963, New York, NY: Springer; 2019, pp. 25–9. https://doi.org/10.1007/978-1-4939-9176-1_4. [16] Green MR, Sambrook J. Isolation of High-Molecular-Weight DNA Using Organic Solvents. Cold Spring Harb Protoc 2017;2017:pdb.prot093450. https://doi.org/10.1101/pdb.prot093450. [17] Edson SM. The effect of chemical compromise on the recovery of DNA from skeletonized human remains: A study of three World War II era incidents recovered from tropical locations. Forensic Sci Med Pathol 2019;15:542–54. https://doi.org/10.1007/s12024-019-00179-2. [18] Holz F, Carrillo-Núñez GG, Martinez Peña EG, Rivera Martinez AA, De La Peña Jiménez IG, Bonilla Virgen R, et al. A guide to classify tattoo motives in Mexico as a tool to identify unknown bodies. Int J Legal Med 2022;136:1105–11. https://doi.org/10.1007/s00414-022-02814-0. [19] Holz F, Carrillo Nuñez GG, Esparza Quezada G, Rivera Martínez AA, Birne V, Schof S, et al. La identificación de personas fallecidas en México – retos actuales y potencial de los tatuajes. Revmedforense 2025;10:21–40. https://doi.org/10.25009/revmedforense.v10i1.3061. [20] Birne VL, Birngruber CG, Vennemann M, Bauer H, Verhoff MA, Quezada Esparza G, et al. Tendons and table salt: A recipe to preserve human DNA. Forensic Sci Int 2024;365:112254. https://doi.org/10.1016/j.forsciint.2024.112254.
Author:
Valentina Leonie Birne, PhD Research Assistent In 2021, I started my Master Thesis on Forensic DNA Phenotyping and the molecular prediction of hair, eye and skin colour at the Institute of Legal Medicine in Frankfurt am Main, Germany. Following my Master’s Degree in Molecular Biosciences, I was lucky enough to work as a research assistant for the German-Mexican cooperation “CoCiMex”, a research project supported by the German Academic Exchange Service (DAAD) on behalf of the German Federal Foreign Office (Project ID: 57594060). Since 2023, I have been working as PhD candidate in the department of Forensic Biology at the Institute of Legal Medicine in Frankfurt am Main, Germany. In my research, I focus on the genetic identification of human remains and work as research assistant for the project “IDH” Identificación Humana en México, a project financed by the Unites Nations Population Fund (UNFPA) with the aim of identifying unknown deceased persons in Mexico.
Images:
Header: Adobe Stock/.rodrigo