5
European descent, even advises that morphological analysis should be used instead of Fordisc for
individuals of certain ancestries until proper reference samples for said ancestries could be added
to the program’s database (20). Validation studies on these methods using remains of modern
context and from various ethnic backgrounds would also provide a way to strengthen these
methods. Texas State University and the University of New Mexico both own decently sized
sample collections of modern contexts that would be suitable for this purpose.
Errors with Fordisc are more prominent in relation to ancestry estimations, especially
when classifying remains of Hispanic descent. An article from 2019 attributes misclassifications
of Hispanic descent to admixture (21). This brings forth another issue within anthropology:
should ancestry estimations even be included in the biological profile? Ancestry is notorious
within the field of biological anthropology for being difficult to assess (22). Due to admixture
and individual variation, ancestry is a much more fluid attribute than the differences between
biological males and females, and this fluidity is in part due to admixture (23). In comparison,
the other three components of the biological profile (age, sex, and height) are relatively fixed
attributes, so the estimations for these categories rely on less ambiguous traits (24). In addition,
ancestry estimations have gained notoriety in articles due to misconceptions of ancestry versus
race (25). Ancestry in anthropology is defined as the geographic region or ancestral origin of a
person, and race is a socially constructed category (22). Although forensic anthropology teaches
and emphasizes this difference by strictly using the term “ancestry,” many researchers argue that
ancestry estimations still create an opening for of engaging in “phenotypic othering” (26).
This phenotypic othering can occur in part due to the difference in terminology and its
implications as used by forensic experts, law enforcement, and lay people. Anthropologists can
estimate ancestry, but this does not imply a direct correlation with skin color. So, even if the
6
estimated ancestry is African, this does not mean the individual is Black. Additionally, since
these are only estimates, the results should not be viewed as absolutes because it can skew
investigations (22,25,26). Despite efforts to distinguish ancestry and race from each other,
misconceptions are still present in literature, forensic cases, and in conversation by the general
public. A language study of the usage of “ancestry,” “race,” and “ethnicity” indicate that the
general public use the three words interchangeably. “Ancestry” was used in the majority of
forensic cases, but it was defined by law enforcement cultural, geographic, governmental, and/ or
self-identity criteria, indicating that ancestry and race were still being viewed as the same.
Interchangeable usage of these terms translates to a perpetual public misunderstanding which
then causes a misrepresentation of forensic research (27). The difficulty of estimating ancestry
combined with society’s misunderstanding of ancestry versus race begs the question of whether
ancestry should continue to be included within the biological profile. Many argue that, because
ancestry estimates have the possibility of hindering identification, it is counterproductive and
should not be used (26). The latest rebuttal to this argument states that it should be reported but
not used by itself. Ancestry estimates based on the skeletal remains should be combined with
other methods such as genetic analysis, and the caveats of these estimates need to be made
abundantly clear to law enforcement and the general public (28, 29).
A similar problem is the divide between sex and gender. Forensic anthropologists are
only able to provide estimations of an individual’s biological sex (30). Since gender is a socially
constructed categorization, forensic anthropologists’ estimation of biological sex does not
provide information to an individual’s gender (31). Juvenile remains are shown to be another
limitation of sex estimations. The traits commonly assessed for biological sex are still developing
and remain ambiguous in juveniles until after puberty (32). Other problems arise from the
7
varying degree of sexual dimorphism between populations and the unknown effects of pregnancy
on ossa coxae morphology (33). Sexually dimorphic traits are not as simple as whether the
observed trait is present or not, but rather, these traits fall within a scale according to
prominence/robusticity. The degree of ruggedness of these traits can be misperceived depending
on the overall size of the individual. Individuals from Asia for example tend to have smaller
builds, and this can make the skeleton seem more gracile than they actually are (34). To avoid
this, sexually dimorphic traits should be considered relative to the size of the individual to
determine prominence/robusticity, and since many sex estimation methods are visual
assessments, interobserver error is another key issue (25). Furthermore, facial surgeries are now
available for individuals who are transitioning as a way to better reflect their gender. Forensic
anthropologists know little about the impact this will have on sex estimates, but preliminary
studies suggest that these surgeries provide enough evidence to make inferences on gender.
However, the majority of these surgeries are feminization surgeries, so this would not be able to
be as widely applied to transgender men (female to male) (30).
MtDNA: Identification and Ancestry Estimation
Inherited solely from the maternal lineage, mitochondrial DNA (mtDNA) was first
discovered by Margit and Sylvan Nass in the 1960’s via electron microscopy (35). The
mitochondrial genome spans approximately 16,500 base pairs and is passed onto offspring
through the maternal lineage (36). Within the mitochondrial genome, 37 genes exist that code for
13 proteins, 22 tRNAs, and 2 rRNAs. The 13 proteins coded for all instruct cells to generate
protein subunits of the enzyme complexes of the oxidative phosphorylation system, enabling the
mitochondria to carry out its function as the powerhouse of our cells (37). The D-loop is a
control region within the genome that is 1,122 base pair long, and it is the most polymorphic
8
region. This region can be divided into the Hypervariable region I (HVI) and Hypervariable
Region II (HVII) (38).
The Cambridge Reference Sequence (CRS) was the first complete human
mitochondrial genome to be sequenced. It was initially published in 1981 but has since been
revised (39). Compared to nuclear DNA (nDNA), mtDNA is both more abundant and prone to
mutation, and overall, mtDNA survives longer than nDNA due to its abundance. Thus, mtDNA
is beneficial in cases where the sample DNA being sequenced is ancient or degraded due to the
high number of copies (40).
Forensic research has spotlighted mtDNA sequencing because of these advantageous
characteristics, and studies have shown that mtDNA sequencing is a robust method regarding
cases of exclusion or absence of identity (36). Since the 1980s, mtDNA has become routinely
used in forensic investigations, and it is considered the last resort for remains or samples that are
highly degraded (41). In 1991, Stoneking et al. published the first report of mtDNA typing being
successfully used in a forensic human identification case. The forensic lab was able to identify
the skeletal remains of a child by using hybridization with 23 sequence-specific oligonucleotide
probes targeting 9 regions of HVI and HVII on the control region. The resulting mtDNA type
matched with the suspected mothers. The conclusions from this case lead to scientists to posit
that mtDNA typing could have potential in sexual assault cases too (42).
Additionally, despite the successful identification in the aforementioned case, mtDNA
analysis is still limited in the results it can provide because it only follows the maternal lineage.
As such, mtDNA has a lower discrimination power when compared to nDNA, which becomes an
issue in forensics when common mtDNA types are involved (36). For example, in Europeans,
the most common HVI/HVII type is found in approximately 7% of the population (43). As a way
9
to increase the discrimination power, researchers have begun large scale sequencing of the entire
mitochondrial genome of individuals with common HVI/HVII types (43, 44).
Data from mitochondrial DNA sequencing can not only be compared for matches but
can also be used as an indicator of lineage. Phylogenetically related haplotypes are formed
through the accumulation of variation (45). Beginning with “Mitochondrial Eve”- the most
recent common ancestor matrilineally- four haplogroups (L0, L1, L2, L3) specific to sub-
Saharan Africa split off and continued to branch out, forming sub-haplogroups (46). For
example, the M and N lineages originated from the L3 haplogroup (47). These haplogroups
continued to branch off as depicted in Figure 2, and the current mtDNA phylogenetic tree
contains more than 4,000 haplogroups (48,49).
The different haplogroups can be categorized by the geographic region according to
their frequency within those regions under the assumption that a high frequency in that area
means that haplogroup likely originated there, but this is not always proven to true (37). With
this assumption though, mitochondrial DNA can be sequenced and matched to a haplotype based
on its specific single nucleotide polymorphisms (SNPs), and from that, the haplotype or
overarching haplogroup can be used to infer maternal ancestry (50). For example, haplogroups J
and U are regarded as part of the European haplogroups because of their high density in that
region. The F haplogroup is predominately considered as an Asian haplogroup because it ranges
from 31 to 77 percent in different parts across Asia (51). The inherent problem with this method
in ancestry assessment is the underlying assumption described previously. Because populations
continue to migrate and new mutations appear in these populations, the difficulty of locating the
origin of a specific haplogroup or mutation increases (37). Therefore, inferences based upon the
method should be considered with some caution.
10
Anatomical Specimens
Often found within schools and skeletal collections, anatomical specimens are deeply
entangled with the sciences and biological anthropology, and one of the first records of
anatomical studies using human remains trace back to the 5th century BCE (52). Use of human
remains in science studies became more pervasive during the 17th century, and in France, the
studied remains were often criminals who had been executed (53). Unlike France, anatomical
specimens in America were not obtained from within the nation. Since America’s mortuary
practice timeline is short, meaning individuals are buried soon after death, the majority of
anatomical specimens present in America originated overseas, and if they were used in studies,
the remains usually ended up on a medical dissection table (54). Americans, due to their fear of
death and the unknown, attempt to deny and defy death through their mortuary practices (55).
Part of this is done by burying the decedent relatively quickly as stated above. On average,
individuals are buried between three to five days after death (56). The other part of this can be
seen through the prevalence of embalming in American mortuary culture. Embalming preserves
the state of the body, and in a sense, disguises all the signs of death by delaying the
decomposition process (57).
As a result of America’s mortuary practice, anatomical specimens had to be
outsourced, and a large percentage of these specimens were imported from India, which had been
a leading producer of human remains throughout the 1970s (58). Since export of human remains
was banned in India during the mid-1980s and China (another big leader in producing anatomical
specimens) in 2008, American retailers have had to purchase the specimens from international
third parties (54, 58). Unfortunately, many of these remains were obtained under unethical
circumstances. A large portion of the medical specimens in India and those that were exported
11
from India were unclaimed bodies. Economic conditions and poor communication modalities
caused a massive collection of unclaimed bodies in India, which the government allowed to be
sold and exported. The number of unclaimed bodies has since declined due to the public outcry
and multiple legislative initiatives being implemented (59).
Recently, activists have been raising attention to the vast amount of remains that had
already been exported under these conditions and has led a call for the repatriation of these
anatomical specimens (54, 58). On top of this, activists are also fighting for a reevaluation of the
human bone trade (58). With the explosion of multiple social media platforms, new avenues for
buying and selling human bones have appeared. In particular, TikTok user Jon Pichaya Ferry has
been the center of controversy because of his collection of human bones. Jon Ferry sells these
bones online under the claim that the bones he sells were all bones specifically prepared to be
used as anatomical specimens (60). However, others have argued saying the origins of these
remains are questionable at best and could have been initially obtained through unethical and/or
illegal means (58). This could occur through situations like in India as previously described,
grave robbing, or even murder. In 1828, William Burke and William Hare became infamous for
having murdered sixteen people in order to sell the bodies to doctors as dissection specimens
(61). While murder is not a common occurrence in this trade, these murders emphasize the
potentially problematic history of anatomical specimens with unknown origins (49).
Applications to BCC #20-100
Case BCC #20-100 provides a key example of why the methods for constructing the
biological profile should be strengthened as well as the importance of anatomical specimens and
their origins. The remains in questions were prepped and used as an anatomical specimen.
However, as mentioned, this does not mean that the individual consented to this prior to death.
12
Providing as much identity as feasible to the individual is one of the only options that can be
done at this point. Reconstructing the identity of the deceased is a significant and serious task, so
it is important that the methods used to do this are as stringent and widely applicable. Accuracy
of these methods need to be determined via large, well-designed studies on individuals of
varying backgrounds. In this case, the biological profile of case BCC #20-100 was constructed
using the methods common in forensic anthropology. However, these methods are based on
specific reference samples that do not match with the traits of this individual. Because of this,
molecular analysis was used to strengthen the ancestry estimation but is limited to only the
maternal ancestral lineage. The features of this case ties into multiple areas of study and their
uses while also underscoring the key problems in these fields.
Materials and Methods
Sex
Sex estimation from the ossa coxae is widely known as having the highest accuracy
largely due to the pivotal role of the pelvic girdle in childbirth, and as a result, the structure of
the human pelvis has multiple sexually dimorphic features that can be used for sex estimation
(63-65). Commonly observed features include the shape of the subpubic angle, presence of a
ventral arc, width of the ischial-pubic ramus, angle of the greater sciatic notch, and shape of the
preauricular sulcus (66-70). Two methods of sex estimation based on the sexually dimorphic
features of the pelvis were conducted using the left os coxa. The Phenice (1969) and Klales et al.
(2012) methods both focus on the pubic portion of the os coxa. The former is a visually based
method while the later employs an ordinal scoring system and linear regression equation to
estimate biological sex (66, 70).
13
Despite the persisting notion that the cranium is the second-best indicator of sex,
multiple studies have determined that the majority of post-cranial markers are more accurate than
the cranial based methods (71-74). As a manner of estimating sex utilizing post-cranial elements,
the morphology of the distal humerus was examined using the Rogers (1999) method, which
indicated a 92% accuracy of sex estimations when performed in the initial study. This manner of
analysis focuses on three main traits of the distal humerus- the shape of the olecranon fossa,
angle of the medial epicondyle, and trochlear extension (75).
Although post-cranial studies showing higher accuracy results, the cranium can also be
used as an indicator for biological sex estimations with approximately a 70% accuracy (76). This
can be especially useful in cases where the ossa coxae are not present or too damaged to be used
(3, 77). A morphometric assessment of the skull was used to estimate sex based on five traits that
are generally recognized as being sexually dimorphic: nuchal crest, mastoid processes,
supraorbital margins, glabella, and mental eminence (78). This method scores each trait on a
range of 1 to 5 using the Buikstra and Ubelaker (1994) scoring and uses a logistic regression
equation to provide an overall percent probability for the estimated biological sex (78, 79).
Age
The ranges for age at time of death was estimated using a variety of methods based on
cranial and post-cranial skeletal markers. Since epiphyses and sutures fuse at different stages
during human development, they can be a highly reliable source of age estimation, particularly
for juveniles (79-82). Moreover, even if a specified age range is unattainable, sutures can be used
to establish whether the bones in question belonged to an adult or not, and based on that result,
one can then choose the next most appropriate method for estimating age. In this case, two
sutures were examined- the basilar suture (79) and medial clavicular epiphyses (83).
14
The use of dentition in age estimations is common practice, and, similar to epiphyseal
fusion, it is especially accurate when working with subadult samples (12, 84). While dental
eruption can help provide foundation for a minimum age estimate, dental attrition is more useful
when working with adult samples (85). Maxillary and mandibular dental wear was assessed
using the Brothwell (1981) and Lovejoy et al. (1985) methods (11, 12).
Sternal rib end analysis was utilized for the age-at-death estimation because it has been
shown to be a highly accurate age assessor in skeletons of both sexes (86). According to Iscan et
al. (1984), rib end characteristics such as pit shape and wall configurations are regarded as being
more accurate than just using pit depth (87). It is cautioned that researchers and anthropologists
consider that sternal rib ends can be influenced by biomechanical variation and sex and can
therefore impact age estimations (87,88). Although research has suggested that sternal rib end
analysis methods can be used on the third rib, the fourth rib was analyzed for this case using the
revised descriptions within the Harnett (2010) study (6). Results were applied to ageing
categories derived from the Maricopa Forensic Science Center (FSC) samples (Harnett 2010)
and the Iscan et al. (1984) samples for males (6, 87).
Estimations using the pubic symphyses were not performed due to the surface of the
symphyses being obscured by the padding that was glued on when the skeleton was re-
articulated after death.
Ancestry
To estimate ancestry, mitochondrial DNA (mtDNA) analysis was executed using bone
samples from the petrous portion of the right temporal bone by using a Dremel rotary tool to drill
through the external auditory meatus (89). Drilling took place in room #237 over a sterilized
15
sheet of butcher paper inside a fume cabinet, and personal protective equipment (PPE) consisted
of gloves that were sterilized with DNA Away, masks, and goggles. Approximately 0.082 grams
of bone dust was collected into a weigh boat and placed in a 2 ml tube. All plastic consumables
had been exposed to UV-C light to decontaminate surfaces for at least 15 minutes.
Extraction was performed using a modified version of the Dabney protocol (90). A
negative control sample was prepped with the same procedures to test for contamination being
introduced from lab reagents. One milliliter of EDTA was added to the extraction tubes to
breakdown the samples, and the tubes was vortexed to ensure proper mixing. The samples were
stored in a drawer overnight. Ten ul Proteinase K was added to the tube on the following day to
lyse the cells. Following this, the samples were allowed to incubate with parafilm sealing the top
to prevent leakage for approximately ~1 hour at 55 degrees Celsius to inactivate the Proteinase
K. After incubation, samples were spun down in a centrifuge, and 1 ml of supernatant was
transferred to a 15 ml tube with 13 ml of PB buffer (Qiagen). This mix was then placed in a
Qiagen spin column and spun through a silica extraction filter. The DNA was bound to the silica
membrane. Afterwards, two 700ml ethanol washes were performed to wash away any
contaminants, and then a dry spin was done. Lastly, two more spins were done using 55 ul of
double distilled water (ddH2O) to elute the purified DNA. The DNA was quantified using a
Qubit (ThermoFisher; 0.245ng/ul) before starting the library preparation protocol, which is the
first step Next-Gen sequencing. Library preparation allows for the DNA to adhere to the flow
cell and for the sample to be identified.
In Dr. Snow’s aDNA lab at the University of Montana where the analysis was
executed, a hybrid of the Meyer-Kircher (2010) and KAPA (Roche) protocols were used for
library prep (91). A master mix (5ul NEBuffer2, 0.25 BSA, 0.5ul ATP, 0.6ul dNTPs, and 1ul
16
water per sample) for blunt-end repair was created. Once added to the samples along with 2ul T4
PNK and 2ul of T4 DNA Polymerase, the samples were incubated at 25 degrees Celsius for 30
minutes in the thermocycler since the DNA ligase enzyme reaches optimal activity at this
temperature. A bead clean was performed, and the beads underwent two ethanol washes (70%
ethanol). Next, 15ul of EB Buffer was added to each sample. Samples were loaded on a magnetic
rack once again to remove the purified DNA solution. The solution transferred to new PCR
tubes, and the DNA was resuspended in 18ul of EB.
To add adapters onto the DNA fragments, an adapter ligation master mix (14.8ul
water, 8ul 5X rapid ligation buffer, 0.2ul annealed adaptors, and 2ul T4 DNA ligase per sample)
was prepped, and 25ul of the mix was aliquoted into each PCR tube along with 15ul of blunt end
repaired DNA. Samples were mixed gently and incubated for 5 minutes in the thermocycler at 22
degrees Celsius. Two bead cleans were performed. To fill in the single-stranded adapter
sequences and complete the adapter annealing process, an adapter fill-in master mix (5ul
ThermoPol Buffer, 0.6ul dNTPs, and 27.4ul water per samples) was created, and 40ul of this mix
was transferred into each of the tubes. Adapter ligated, purified DNA (15ul) and Bst DNA
Polymerase (2ul) were also added to each tube. Samples were mixed and incubated again at 37
degrees Celsius for 30 minutes and then at 80 degrees Celsius for 20 minutes.
Prior to another bead clean and to enable pooling of samples, 40ul of an indexing
master mix (5ul ThermoPol Buffer, 0.5ul dNTPs, 1ul AmpliTaq Gold NDA Polymerase, 33.5ul
water per sample) along with a P5 adapter (5ul), P7 adapter (5ul), and DNA from the adapter fill-
in step (~50ul) were aliquoted into each tube. Following this, amplification of the sample library
using LM-PCR began as described by KAPA. Thirty microliters of a Pre-Capture LM-PCR
master mix was added to each sample. The PCR tubes containing the samples were then placed
17
in the thermocycler for amplification following the Pre-Capture LM-PCR program. The resulting
amplified library was purified via a bead clean. Qubit was used again to quantify the DNA
(52ng/ul).
Samples were then pooled along with aDNA samples from the Molecular
Anthropology summer course during which this analysis occurred. After a bead clean,
hybridization probes were added. Hybridization incubation occurred overnight at 95 degrees
Celsius for 5 minutes then at 47 degrees Celsius for 20 hours. The DNA was captured using a
bead wash. The beads with the bound DNA were washed using 40ul Stringent Buffer, 30ul Wash
Buffer I, 20ul Wash Buffer II, 20ul Wash Buffer III, and 1000ul Bead Wash Buffer. The Post-
Capture LM-PCR program was used to amplify the bead bound DNA. The samples were purified
via a bead clean, and the DNA was quantified again.
Finally, samples were sequenced at the Genomics Core at the University of Montana
using Illumina MiSeq Sequencing. Sequencing data was uploaded to a web-based bioinformatics
platform (Galaxy)- using a pipeline created for Dr. Snow’s aDNA lab. The raw data was
inputted, and a FastOc was run to perform basic quality control checks. The pipeline then
trimmed the adaptors and merged the forward and reverse reads. Next, the reads were mapped to
a reference genome, and Qualimap generates data for read depth, sequence coverage, and the
number of reads. Genotype calls then a consensus file are generated, and the overall sample
contamination was estimated. The finished data analysis in Galaxy was entered into Haplogrep, a
database that simplifies the process of identifying mitochondrial haplogroups by using pre-
calculated phylogenetic weights that correspond to the occurrence per position in Phylotree (92).
Results were organized and compiled into a spreadsheet.
18
In conjunction with the DNA analysis, non-metric and metric analyses were both used
to estimate ancestry from the cranium. Non-metric cranial analysis includes a visual based
method that uses a chart which classifies the different variants of a specific craniofacial trait into
geographic races based on observed frequency (93). Additionally, another visual based method
that uses a scoring system for the cranial traits was utilized (94). Based on the scores, the Hefner
software will create probability matches for each ancestry. Ancestry based on craniometric data
(Figure 3) was analyzed via the Fordisc 3.0 program (95). Cranial measurements were analyzed
in Fordisc using the FDB dataset. The data was processed using all male populations available,
and results were based on 21 measurements. Measurements with high deviations and were
removed, and results were processed again. The input data was whittled down further to improve
overall accuracy of the results by removing reference populations that were highly dissimilar.
Stature
Stature is a population dependent analysis, and since the individual for this case shows
a mixture European and Asian characteristics, the Trotter (1970) method was utilized because it
provides equations for both of these populations (96, 97). Equations for European and East Asian
males were calculated using measurements from the radius, femur, and fibula. Certain
measurements from the ulna and tibia were excluded because they could not be disarticulated
without further damage to the bones. Therefore, accurate measurements could not be taken for
these elements (Figure 4).
Although the Fordisc 3.0 program does not provide an Asian reference population for
stature estimations, it does provide a European reference population. As such, Fordisc 3.0 was
still used because the individual exhibits a mixture of characteristics. Stature was assessed using
both European and African male reference populations using Trotter statistics to provide a more
19
inclusive estimate. Furthermore, a 95% confidence probability was used for these calculations.
Once again, certain measurements were excluded from the ulna and tibia because the remains
could not be fully articulated without further damaging the bones. Due to post-mortem damage,
pelvic heigh and ischium length were excluded too.
Trauma and Pathology
The Skeletal Trauma Analysis: Case Studies in Context (98) was used to guide the
observation and analysis of trauma for this case. For possible skeletal pathologies, the Human
Bone Manual (99) and the Identification of Pathological Conditions in Human Skeletal Remain
(100) were used as reference.
Biological Profile BCC #20-100
Sex
Os Pubis Dimorphic Traits
The Phenice (1969) method is a visual assessment of biological sex based on the
sexually dimorphic features of the ventral arc, subpubic concavity, and medial aspect of the
ischiopubic ramus (66). The os coxa exhibited the following: minimal presence of a ventral arc,
little to no subpubic concavity, and a broad ischiopubic ramus with no distinct ridge. These
features are most commonly observed in males, indicating the individual was male.
The Klales et al. (2012) method was developed to review and revise the Phenice
(1969) method, and it provides a percent probability report for the estimated sex. The same three
traits are used and are scored on a range from 1 to 5, with 1 being the extreme traits on the
female side and 5 being the extreme for the male side of the spectrum (70). The ventral arc,
20
subpubic concavity, and medial aspect of the ischiopubic ramus each scored as a 4, indicating a
98% probability the individual was male (Figure 5).
Morphometric Skull Assessment
Out of the 5 traits examined for the Buikstra and Ubelaker (1994) method, the
glabella/supraorbital ridges, mastoid processes, and nuchal crest all scored at a 3 (79). Both the
mental eminence and orbital margins scored at a 4, exhibiting slightly more robusticity than the
other features. These scores indicate with an averaged 82% probability the individual was a male
(Figure 6).
Distal Humerus Morphology
The general shape of the olecranon fossa, angle of the medial epicondyle, and
trochlear extension were employed as indicators for sex (75). For case #20-100, the individual
presented olecranon fossae that were roughly triangular shaped. Additionally, the medial
epicondyles were parallel or nearly parallel to a table’s surface when held with the posterior end
up, and the medial edge of the trochlea extended farther than that of the lateral edge. All 3 traits
are most frequently observed in men, and as such, are indicative of that the individual was male
(75). Results from each of the aforementioned methods indicate the individual was male.
Age
Suture Closure
The individual’s basilar suture is completely fused, indicating a minimum age of 23
years (79). Likewise, the medial clavicular suture was also completely fused. This correlates with
a minimum age also of 23 too according to samples retrieved from male U.S. military personnel
21
who died during the Korean War (83). Thus, it can be established that the individual was an adult
at time of death.
Dentition
Dental attrition for case #20-100 was estimated in reference to the degree and patterns
of wear as depicted in the dental attrition charts for Lovejoy et al, (1985) and Brothwell (1981).
Maxillary dental attrition patterns aligned most closely with the phase E classification,
correlating with an age range of 24 to 30 years of age (Photo 9). Furthermore, mandibular wear
resembled that of phase G, which correlated to a slightly later age estimate of 35 to 40 years of
age (12; Photo 10). The two estimates provide an overall age range of 24 to 40 years old;
however, this method was developed based on samples from a pre-historic American Indian
population from Libben, Ohio. Therefore, the population and antiquity of the samples for this
method must be noted when taking these results into consideration.
To create a more inclusive and well-rounded age estimate, the Brothwell (1981)
method was applied as well (11). Maxillary and mandibular molar wear patterns both matched
best with a score of 3+ to 4. This has an associated age range of 25 to 35 years of age. It should
be noted that this analysis is based solely on the first and second molars, and all the third molars
were not included because they appear to be congenitally missing. As mentioned above, some
discretion is advised regarding the age estimate results since this method is based on medieval
English skeletons.
Sternal Rib End Analysis
The left and right rib ends both had shallow, V-shaped pits that did not show signs of
billowing (Photo 11 and 12). The rims were smooth and round, without any scalloping or
22
irregular flaring. Minimal microporosity was present on the left rib pit. Rib end observations
corresponded the best with the description for Phase 2. Phase 2 has an associated age range of 18
to 25 years based on the European rib end samples from the Iscan et al. (1984) study and 21-28
based on the FSC samples used in the Harnett (2010) research (6, 87).
Auricular Surface
Age estimates from the auricular surface can be broad, and the mean ages for some of
the methods do not differ significantly between each of the phases. However, since the auricular
surface changes in a regular manner throughout life, it can be argued that it is more reliable than
the pubic symphyses (101). For a comprehensive estimate, the Lovejoy et al. (1985) and the
Chamberlain and Buckberry (2002) methods were both applied to the left and right auricular
surfaces (102, 103). The later method is often regarded as the more accurate method because it
uses a scoring system rather than just phase descriptions (104).
Topography of the auricular surface was observed and cross-referenced with the eight
phase descriptions provided by Lovejoy et al. (1985) (103). Since striae were still faintly present
on the superior demi-face as well as smooth granularity, the individual placed within phase 3 for
the left and right, suggesting an age range of 30 to 34 years of age (Photo 13). Additionally,
minimal retro-auricular activity was present, and the individual exhibited little to no apical
changes. Moreover, a scoring method was used based on five characteristics: transverse
organization (3), surface texture (4), microporosity (2), macroporosity (1), and apical changes (2)
(102). With a combined score of 12, the associated stage is stage IV, which provides an
estimated age range of 37 to 65 years of age and a mean age of approximately 51 years old.
Combining all the results for age estimation, the overall age range of the individual is 20 to 40
years old with a most likely range of 25 to 35 years of age.
23
Ancestry
Mitochondrial DNA
The Miseq results from Galaxy and Haplogrep are provided in Table 1 which includes
data for the following categories: percent endogenous reads (total mapped unique reads), average
read length (fragment length measured in base pairs), depth of reads (number of times each base
pair was sequenced), percent of reference sequenced covered >1x and >2x (coverage of the base
pairs used for haplogroup assignment), variant sites with 1x and >1x coverage (sites that vary
from the reference that are only sequenced once and the ones sequenced more than once), and
the haplogroup result. Sequencing results show a 179X read depth with a 100 percent coverage.
Fragment misincorporation plots, or damage plots, are also provided for the forensic sample and
control to indicate the degree of degradation which correlates with the age of the samples. The
modernity can be loosely approximated based on the degree of damage represented in these
plots. The damage plot for sample BCC #20-100 depicts a red line (C→T substitutions) and a
blue line (G→A substitutions) that curve slightly at the ends while laying relatively flat in the
center of the plot (Figure 7). The degree of curvature at the ends of these lines indicated the level
of damage to the DNA, with larger curves representing more degraded samples in comparison to
the reference, and flatter line representing fewer degraded bases (105). The present curvature
does indicate the DNA was damaged but not to the extremes that would indicate the antiquity of
the skeleton is ancient. Figure 8 depicts the damage plot for the negative control sample. The
horizontal lines intermittently interrupted by spikes in the map indicate the presence of trace
amounts of DNA. This suggests sample contamination from short fragments of DNA did occur
but only to a slight degree, which is to be expected.
24
Additionally, results indicated the individual belonged to the haplotype J1c3f. The
haplogroup J it is most frequently found in Europe (106). However, the J haplogroup has been
shown to have some frequency (12%) in the Near East. The most common hypothesis is that, like
most European haplogroups, the J haplogroup originated in the Near East and was brought to
Europe by the first farm-herding societies during the Neolithic period (107). Geographic origins
of the main mtDNA haplogroup based on Mancuso et al. (2008) places the J haplogroup in
Western Europe as depicted in Figure 2 (48). Overall, the mtDNA results indicate a European
maternal ancestry for the individual based on the geographic frequency of the J haplogroup.
Paternal ancestry is not accounted for in this analysis and could be distinct.
Non-Metric Cranial Assessment
Observed non-metric craniofacial traits include a medium nasal spine, moderate nasal
aperture width, absence of a post-bregmatic depression, an angled zygomaticomaxillary suture,
and wide mastoid form. In reference to the Gill (1986) chart, the individual matched the most
with trait descriptions of Eastern Asians, Polynesians, and American Indians. Not all features in
this chart were assessed. Ambiguous and damaged features were excluded. As presented in
Figure 9, cranial trait scores analyzed based on the Hefner (2009) method showed a higher
probability of American Indian (.4) ancestry followed closely by Asian (.32) and then European
(.22). Mitochondrial DNA analysis and skeletal based methods show a mixture of European and
Asian ancestry.
FORDISC 3.0
After excluding measurements with high deviations and the least likely reference
populations, the final calculations were based on 17 measurements and 5 reference groups
25
(White males, Chinese males, Black males, Vietnamese males, and Guatemalan males). The
results indicated the individual was most similar to Chinese males with a posterior probability
of .785 and typicality of .013 (Figure 10). White males were the next closest population match
but had a significantly lower posterior probability of .148.
Each of the methods for ancestry estimation had differing results, but overall, showed
a mix of European and Asian characteristics. Considering the remains were used as an
anatomical specimen and the history of anatomical specimens as mentioned above, it is likely
that the individual was from India.
Stature
Trotter (1970) Equations
Calculations from the European male equations for the radius, femur, and fibula
provided an overall range of 155.8 to 167.5 cm, which is equivalent to a range of 5’1” to 5’6”.
Results based on East Asian males gave an estimated range of 154.9-169.3 cm (5’1” to 5’6”).
The overall stature range per this method is 5’6”.
FORDISC 3.0
Similar to the Trotter (1970) method, specific measurements from the ulna, and tibia
were excluded. Two measurements from the ossa coxae, pelvic height, and ischial length, were
also excluded due to damage on the ischial tuberosities. Results from the Fordisc 3.0 analysis
(Figure 11) indicated an estimated stature between 61.3 to 66.7 inches (5’1” to 5’6”). The short
regression formula for the femur and fibula provided an estimate of 5’4” with a standard
deviation of 2.2”, which corresponds with an approximated range of 5’1” to 5’6”. Per the Fordisc
3.0 results, the overall estimated stature for the individual is between 5’1” to 5’6”.
26
Trauma and Pathology
The majority of trauma present on the skeleton is due to post-mortem events, including
trauma resulting from articulation of the skeleton, normal wear from use as an anatomical
specimen, exposure to fluctuation temperatures, and the disarticulation process. A hanging hole
is present and located slightly anterior to the bregma (Photo 14). Other drill/bolt holes are present
on the glenoid fossas, sacrum, and ossa coxae. The maxilla and mandible exhibit cracking and
flaking, plaster and glue is present where someone in the past tried to fix the fracturing areas
(Photo 15). The left and right ribs 2 through 6 are fractured vertically along the shafts (Photo 16),
and the medial ends of the clavicles (Photo 17). Near the 5th segment, the lateral edges of the
sacrum (Photo 18) show minimal signs of damage. As previously mentioned, during the
disarticulation process, the posterior wall of the pubic symphyses were damaged.
One area of ante-mortem trauma was noted on the left 11th rib which was identifiable
by the presence of a bony callus (Photo 19), indicating healing had begun before time of death
(98). No major or identifying pathologies were noted.
Conclusions
A comprehensive forensic case analysis was performed for case BCC #20-100 that
employed the use of multiple methods to estimate sex, age, ancestry, and stature in conjunction
with associated antemortem and postmortem trauma. Results indicate the individual was likely a
male between 25 to 35 years of age with an estimated stature range of 5’1” to 5’6”. Skeletal
features show a mix of Asian and European characteristics, and it is likely the individual was
from India. No major identifying pathologies were found. The only location of antemortem
trauma noted was on the left 11th rib and is identifiable by a bony callus on the midshaft.
27
Appendix
Table 1. Condensed Galaxy and Haplogrep results.
Sample
%
Endogenous
reads
Avg.
length
% Ref-
seq
covered
>1x
% Ref-
seq
covered
>2x
Variant
sites 1x
coverage
Variant
sites >1x
coverage
Mean read
depth
Haplogroup
20-100
4.6%
92.64
100%
100%
0
25
178.9915x
J1c3f
Control
10.1%
77.73
18.69%
1.83%
9
0
.2111x
H2a2a1
Figure 1. Dental chart illustration. Shaded areas represent absent dentition.
28
Figure 2. Haplogroup migration routes (Mancuso et al, 2008).
Figure 3. Cranial measurements. The asterisks indicate measurements later discarded because of
unreliable accuracy due to post-mortem damage.
29
Figure 4. Post-cranial measurements. The asterisks indicate measurements later discarded
because of unreliable accuracy due to post-mortem damage.
Figure 5. Klales et al. (2012) os coxa sex estimation trait scores and results.
30
Figure 6. Walker (2008) cranial sex estimation trait scores and results.
Figure 7. Fragment misincorporation plot for BCC #20-100.
31
Figure 8. Fragment misincorporation plot for the negative control.
Figure 9. Hefner (2009) cranial ancestry estimation results. Probability outputs from left to
right: .07, .40, .32, .22.
32
Figure 10. Fordisc 3.0 ancestry estimation results based on cranial metrics.
Figure 11. Fordisc 3.0 stature estimation results based on post-cranial measurements. Trotter M
Stats were used as reference with a 95% confidence interval.
33
Photo 1. Remains of BCC #20-100 as received by the Bonner County coroner.
34
Photo 2. Anterior view of the sternum with part of the remaining cartilage.
Photo 3. Post-mortem damage to the left ischial tuberosity.
35
Photo 4. Post-mortem damage to the right ischial tuberosity.
Photo 5. Mid-disarticulation process of the spinal column and pelvic girdle.
36
Photo 6. Posterior wall of the pubic symphyses that broke off during disarticulation.
Photo 7. Articulated left humerus and ulna from posterior view.
Photo 8. Articulated left patella, tibia, and talus from anterior view.
37
Photo 9. Maxillary dentition.
Photo 10. Mandibular dentition.
38
Photo 11. Sternal end of the left 4th rib.
Photo 12. Sternal end of the right 4th rib.
39
Photo 13. Right auricular surface.
Photo 14. Hanging hole on the superior portion of the frontal bone, slightly anterior to bregma.
40
Photo 15. Cracking on the left side of the mandibular body.
Photo 16. Ribs laid out in anatomical position.
41
Photo 17. Left clavicle with slight post-mortem damage to the lateral end.
Photo 18. Anterior view of the sacrum.
42
Photo 19. Ante-mortem fracture healing on the left 11th rib shaft.
