Ancient Mammalian and Plant DNA from Late Quaternary Stalagmite Layers at Solkota Cave, Georgia

Abstract

Metagenomic analysis is a highly promising technique in paleogenetic research that allows analysis of the complete genomic make-up of a sample. This technique has successfully been employed to archaeological sediments, but possible leaching of DNA through the sequence limits interpretation. We applied this technique to the analysis of ancient DNA (aDNA) from Late Quaternary stalagmites from two caves in Western Georgia, Melouri Cave and Solkota. Stalagmites form closed systems, limiting the effect of leaching, and can be securely dated with U-series. The analyses of the sequence data from the Melouri Cave stalagmite revealed potential contamination and low preservation of DNA. However, the two Solkota stalagmites preserved ancient DNA molecules of mammals (bear, roe deer, bats) and plants (chestnut, hazelnut, flax). The aDNA bearing layers from one of the two Solkota stalagmites were dated to between ~84 ka and ~56 ka BP by U-series. The second Solkota stalagmite contained excessive detrital clay obstructing U-series dating, but it also contained bear bones with a minimum age of ~50 BP uncalibrated years and ancient DNA molecules. The preservation of authentic ancient DNA molecules in Late Quaternary speleothems opens up a new paleogenetic archive for archaeological, paleontological and paleoenvironmental research.

Full text

1
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
Ancient Mammalian and Plant DNA
from Late Quaternary Stalagmite
Layers at Solkota Cave, Georgia
M. C. Stahlschmidt1,2, T. C. Collin3, D. M. Fernandes4,5, G. Bar-Oz6, A. Belfer-Cohen7, Z. Gao8,
N. Jakeli9, Z. Matskevich10, T. Meshveliani9, J. K. Pritchard8,11,12, F. McDermott13 & R. Pinhasi
4
Metagenomic analysis is a highly promising technique in paleogenetic research that allows analysis
of the complete genomic make-up of a sample. This technique has successfully been employed to
archaeological sediments, but possible leaching of DNA through the sequence limits interpretation. We
applied this technique to the analysis of ancient DNA (aDNA) from Late Quaternary stalagmites from
two caves in Western Georgia, Melouri Cave and Solkota. Stalagmites form closed systems, limiting the
eect of leaching, and can be securely dated with U-series. The analyses of the sequence data from the
Melouri Cave stalagmite revealed potential contamination and low preservation of DNA. However, the
two Solkota stalagmites preserved ancient DNA molecules of mammals (bear, roe deer, bats) and plants
(chestnut, hazelnut, ax). The aDNA bearing layers from one of the two Solkota stalagmites were
dated to between ~84 ka and ~56 ka BP by U-series. The second Solkota stalagmite contained excessive
detrital clay obstructing U-series dating, but it also contained bear bones with a minimum age of ~50 BP
uncalibrated years and ancient DNA molecules. The preservation of authentic ancient DNA molecules in
Late Quaternary speleothems opens up a new paleogenetic archive for archaeological, paleontological
and paleoenvironmental research.
Ancient DNA (aDNA) genomics is a valuable information source on past biological diversity and evolutionary
trajectories of species1–3. A particular focus has been on the analysis of human bones yielding high coverage
genomes of archaic humans4–6 and enabling novel insights into human dispersals and migrations7–9. Additionally,
several studies employed a metagenomic approach to the study of DNA sequence data retrieved from soils and
sediments from various environments, including caves10, lakes11, arid12 and arctic environments13,14. Slon et al.15
using a shotgun sequencing approach and analysing the deamination pattern for identication of authentic
ancient DNA16, reported on the recovery of archaic human aDNA as well as other mammalian aDNA from
archaeological deposits at several sites. is metagenomic research shows that not only bones but many other
components of the archaeological and paleontological record, such as deposits themselves, may serve as a preser-
vation medium for ancient DNA.
e retrieval of authentic aDNA strands from deposits is made possible by the binding of DNA to various
sediment and soil components, including clays17–19, silica20,21, humic acids22 and calcite23. However, soil chem-
istry, e.g. pH20, and soil transformation processes, such as the dissolution and precipitation of minerals, greatly
impacts preservation. Furthermore, post-depositional movement of sediment components through turbation,
such as bioturbation, as well as other soil translocation processes, such as clay illuviation, may negatively impact
the integrity and complicate the interpretation of aDNA found in sediments and soils24,25.
1Department of Human Evolution, Max-Planck-Institute for Evolutionary Anthropology, Leipzig, Germany. 2School
of Archaeology, University College Dublin, Dublin, Ireland. 3School of Medicine, University College Dublin, Dublin,
Ireland. 4Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria. 5CIAS, Department of Life
Sciences, University of Coimbra, Coimbra, Portugal. 6Zinman Institute of Archaeology, University of Haifa, Haifa,
Israel. 7Institute of Archaeology, The Hebrew University of Jerusalem, Jerusalem, Israel. 8Department of Genetics,
Stanford University, Stanford, USA. 9Department of Prehistory, Georgian State Museum, Tbilisi, Georgia. 10Israel
Antiquities Authority, Jerusalem, Israel. 11Departments of Biology, Stanford University, Stanford, USA. 12Howard
Hughes Medical Institute, Stanford University, Stanford, USA. 13School of Earth Sciences, University College
Dublin, Dublin, Ireland. Correspondence and requests for materials should be addressed to M.C.S. (email: mareike_
stahlschmidt@eva.mpg.de) or R.P. (email: ron.pinhasi@univie.ac.at)
Received: 17 January 2019
Accepted: 15 April 2019
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved

2
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
Speleothems are another potential source for aDNA and have long been explored as paleoenvironmental
archives using other methods, mainly stable isotopes studies and U-series dating26. Paleoenvironmental studies
most commonly use stalagmites, which form on the cave oor below a drip and in which calcite precipitates in
distinctive and continuous layers. Preservation conditions for DNA are ideal inside the stalagmites and especially
for those located deeper inside caves, where low temperatures limit the production of reactive oxygen species27,
there is little exposure to UV light28 and a stable pH as well as very low permeability inside the stalagmite and
hence a low risk of DNA migration between consecutive stalagmite layers. Few DNA studies have been conducted
on speleothems and they are mainly restricted to the surface of speleothems29–31 with the exception of a study by
Zepeda Mendoza et al.32, who analysed two samples from the inside of popcorn calcite from a dolerite granite
gneiss cave. However, while they reported that aDNA was preserved inside the speleothems, they concluded that
this type of speleothem is unsuitable as a biological paleoarchive32. ‘Popcorn’ calcite exhibits rather irregular
and complex multi-dimensional growth patterns compared with the relatively simple sequential deposition of
consecutive layers in stalagmites, making the latter a geometrically simpler and therefore more reliable archive.
We here present a rst metagenomic study exploring aDNA metagenomics combined with U-series dates of
stalagmites from two caves from Western Georgia, Solkota and Melouri Cave, as archives on species that inter-
acted with or inhabited these cave systems. In 2016, we surveyed six caves in the Imereti region of Georgia (Fig.1):
three archaeological cave sites - Satsurblia Cave33, Dzudzuana34, Kotias Klde35 - and three non-archaeological
cave sites - Melouri Cave, Datvi Cave, Solkota. e latter three caves contained cave bear bones, but were not
archaeologically explored, and only these sites had favourable speleothems for the aim of this study. Each of the
three archaeological cave sites had a large entrance, permitting light and air to enter into the cave, which typically
makes them less suitable for quantitative paleoenvironmental reconstructions based on stable isotope studies26.
We therefore chose to proceed in our analysis with one stalagmite from Melouri Cave (MEL) and with two stalag-
mites from Solkota (SKK) (Fig.2). Ancient DNA was detected in several locations inside the two Solkota stalag-
mites (SKK 16 3 and 5). However, analysis of the stalagmite from Melouri Cave revealed potential contamination
and low preservation of DNA (see results below, SI Text 1 and Fig.1) and we focus here on the Solkota samples.
Solkota cave lies near the village of Kumistavi above the river Semi. Solkota is part of the same karst system as
Satsurblia Cave, Melouri Cave and Datvi Cave, the Tskaltubo karst system in the Sataplia-Tskaltubo Limestone
Massif36. e cave entrance of Solkota is located in a sinkhole with a very steep slope and little light penetrat-
ing into the cave entrance. e rear of the cave consists of a steep, muddy slope leading upwards with bedrock
exposed at the top of the slope. Another, former entrance may have been present here. Next to limestone boul-
ders the cave contains clay-rich mud and water concentrated in ponds and rills. e cave is rich in speleothems
(stalagmites, owstone, stalactite, curtains, straws) as well as in bone and we also found three lithic akes. Bones
are oen exposed in rill beds and we collected 40 bones. One bone was identied as capra, two as canids and the
remaining 37 as cave bear (Ursus spelaeus or Ursus deningeri). We also observed several bear hibernation dens in
the inner parts of the cave. We collected one large stalagmite, which had been growing on top of three cave bear
long bones (SKK 16 5) from the secondary context of a rill bed (Fig.2B,C). e bones comprised of right and le
distal humerus and a distal sha of a tibia. Carnivore gnaw marks were observed on the surface of the bones. We
collected a second stalagmite (SKK 16 3) from the top of the slope at the rear of the cave, close to its potential root
(Fig.2A).
Results
U-series. Uranium concentrations in speleothem SKK 16 3 are relatively low, typically in the range 35–70
ppb (Table1). ree samples from SKK 16 3 (3/10, 3/6.5, 3/5) (Fig.3) have high 232 contents (c. 42–77 ppb)
resulting in low (230/232) values (between 2.1 and 3.2), and therefore unacceptably large age uncertainties
aer corrections for detrital thorium have been applied (SI Table1). Similarly, the 232 contents for all samples
from SKK 16 5 (Fig.4) are too high to calculate ages for this speleothem (SI Table1). However, six samples from
Figure 1. Location of the study sites. e studied cave sites are located in Western Georgia (map created with
ASTER GDEM59): (1) Location of Satsurblia, Solkota, Melouri and Datvi Cave; (2) Location of Dzudzuana and
Kotias Klde.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

3
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
SKK 16 3, corresponding to d (depth from top) values of 17.5, 13.2, 6, 5.5, 4 and 1.7 cms, have low 232 contents
yield moderately high (230/232) ratios in the range 14.7–148 (Table1), permitting the calculation of precise
U-series ages following correction for the detrital clay component. As discussed in the Methods section, clay-
rich samples from the cave were measured separately using a total dissolution approach to constrain the actual
(230/232) value of the detrital component in the speleothems, considerably reducing the uncertainties in the
corrected U-series ages compared to the standard approach of simply assuming a (230/232) value for the detri-
tal component. Overall, detrital corrected U-series dates for the key stalagmite SKK 16 3 from Solkota range from
83.79 ± 0.64 ka at a depth from top (d) of 13.2 cm to 50.02 ± 0.68 ka at a d of 1.7 cm (Fig.3). In detail however,
considerable complexity in the speleothem’s growth history is evident.
e date from the sample closest to the base of the speleothem (17.5 cm d) yields an age of 80.26 ± 1.87 ka,
out of stratigraphic order, and just outside the error limits of the next three dates above (83.79 ± 0.64, 84.57 ± 0.76,
83.32 ± 1.48 at ds of 13.2, 6 and 5.5 cms respectively, Fig.3). is may indicate some minor post-depositional
migration of uranium in the lower section of the speleothem. Regardless, the similarity of the next three dates (all
three within their 2σ errors) indicates an interval of very rapid speleothem growth around 84 ka and no detectable
post-depositional uranium migration. Warm, wet intervals favour high speleothem growth rates and we note that
this time interval coincides with climatic amelioration during Greenland Interstadial 21.1e (GI-21.1e)37 during
Marine Isotope Stage (MIS) 5a.
Figure 2. Sampling locations of the Solkota Cave stalagmites SKK 16 3 and 5 (photos taken by MCS). (A) e
nd spot of SKK 16 3 (red circle) next to its possible root (white arrow). Note the scarcity of sediment here.
(B) Discovery location of SKK 16 5 (red circle) in a rill bed next to multiple bone remains (blue dots). (C)
Stalagmite SKK 16 5 with cave bear bones at its base.
Sample 238U ppb (230/238U) (234U/238U) (230/232) 232 ppb Age ka
uncorrected Age ka
corrected
SKK16 3/17.5 37.083 ± 0.003 0.6317 ± 0.0020 1.1221 ± 0.0009 14.70 ± 0.04 4.8733 ± 0.0043 88.63 ± 0.54 80.26 ± 1.87 1.84
SKK16 3/13.2 35.263 ± 0.004 0.6138 ± 0.0019 1.1245 ± 0.0014 148.10 ± 0.41 0.4466 ± 0.0030 84.59 ± 0.55 83.79 ± 0.64 0.63
SKK16 3/6 35.34 ± 0.02 0.6140 ± 0.0017 1.1087 ± 0.0012 59.414 ± 0.153 1.1162 ± 0.0003 86.64 ± 0.50 84.57 ± 0.76 0.75
SKK16 3/5.5 64.113 ± 0.005 0.6338 ± 0.0011 1.1156 ± 0.0010 19.086 ± 0.031 6.5056 ± 0.0012 89.95 ± 0.36 83.32 ± 1.48 1.46
SKK 16 3/4 53.907 ± 0.003 0.5001 ± 0.0008 1.1546 ± 0.0007 17.157 ± 0.026 4.8015 ± 0.0012 61.07 ± 0.18 56.69 ± 0.95 0.94
SKK16 3/1.7 38.183 ± 0.005 0.4510 ± 0.0009 1.1658 ± 0.0018 23.266 ± 0.047 2.2618 ± 0.0004 52.72 ± 0.24 50.02 ± 0.68 0.67
Table 1. U-series data for speleothem SKK 16 3. Parentheses denote activity ratios. Dates reported in this table
are considered reliable aer detrital corrections have been applied (see SI Table 1 for dates strongly aected
by detrital correction and with no reliable age calculation). e following decay constants were used: 230:
9.1577E-6, 232: 4.9475E-11, 234U: 2.826E-6, 238UE 1.551E-10. e nal column on the right hand side shows
the ages calculated aer correction for detrital thorium using a measured (230/232) value of 0.95 ± 0.1 for the
detrital end-member.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

4
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
Examination of the cut surface of stalagmite SKK16 3 reveals the presence of three distinctly visible deposi-
tional hiatuses at ds of 4.5 (hiatus 1), 4.8 (hiatus 2) and 5.6 cms (hiatus 3) (Fig.3). ese provide clear evidence
that the speleothem growth was discontinuous above the interval dated at 83.32 ± 1.48 ka. e DNA sample SKK1
is located between hiatus 3 and the combined hiatus 1 and 2. Consequently, the reliable bracketing ages for SKK 1
are 84.57 ± 0.76 ka at 6 cms d (older layer), 83.32 ± 1.48 ka at 5.5 cms d (same layer) and 56.7 ± 0.95 ka at 4 cms
d (younger layer) (Table1, Fig.3). e latter date corresponds to a warm MIS3 interval in the N. Hemisphere
(GI-16.1). e DNA sample SKK3 was taken from the same spot as the u-series sample at 17.5 cm d (Fig.3) with
an age of around 80.26 ± 1.87 ka and is capped by the u-series age 84.57 ± 0.76 ka at 6 cms d (Table1).
Radiocarbon. A fragment of bone from the bottom of stalagmite SKK 16 5 was sent for AMS radiocarbon
dating at the Research Laboratory for Archaeology and the History of Art, University of Oxford. e age of the
bone is beyond the range of radiocarbon, giving it a minimum age of 50.200 BP uncalibrated (OxA-36539).
Ancient DNA. All samples were aligned to the human reference genome (GRCh37/hg19) and damage pat-
terns were assessed. Alignments to the human genome were either too short, <35 bp (base pairs), and aligned
uniquely to the human genome or they were longer, >75 bp, and showed a low deamination rate, indicative of a
high likelihood of contaminant modern human DNA (SI Fig.2). As such, all primate sequences were excluded
from further study due to potential for human contaminant DNA.
Analysis of the Melouri cave samples (MEL1–4) showed that the majority of aligned reads fell within the
ranges of <35 bp, prone to misalignments, and >75 bp, with low deamination indicating potential contamination
and low preservation of DNA of ancient origin (SI Table2). e Melouri samples were therefore excluded from
Figure 3. e cut stalagmite SKK 16 3 (photos taken by MCS). (A) SKK 16 3 before sampling. e stalagmite
was partially cut open with a rock saw and then broken open (broken surface is to the right of the dashed
blue line) to reduce contamination by the saw blade. ree dark lines stemming from hiatuses in speleothem
formation can be observed at ds of 4.5 (h1), 4.8 (h2) and 5.6 (h3) cms (black arrows). Note that hiatuses
h1 and h2 combine to the right (h1, 2). (B) SKK 16 3 aer sampling for DNA analysis and U-series dating.
U-series samples (red, dotted line if unsuccessful analysis) and samples for DNA analysis (green, dotted line
if unsuccessful analysis) were oen taken in close association. Reliable U-series ages are reported next to their
sampling location. Note however, that the age of 80.26 ± 1.87 ka in the same sampling locality as SKK3 is less
reliable as it is out if stratigraphic order. DNA sample SKK1 was taken in the same layer as U-series sample
SKK 16 3/−5.5, between hiatuses h3 and the combined hiatus h1 and h2 and dating to 83.32 ± 1.48 ka. Its age
is capped by U-series ages from layers above (56.7 ± 0.95 ka) and below (84.57 ± 0.76). SKK1 may contain dust
particles from the hiatus events.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

5
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
further analysis. In contrast, almost all Solkota cave samples (SKK1, 3–12) showed preservation of aDNA, most
with multiple genera identications (Fig.5). In the case of SKK 2, screening prior to sequencing showed no dis-
cernible presence of DNA and this sample was therefore excluded from further analysis.
An initial global alignment to the Blastn database with MGmapper revealed 16 commonly occurring genera
in the Solkota samples (SI Table2). e reassessment for false positives (following the approach by Slon et al.15
and see method section below), positively identied 6 genera: Capreolus (roe deer), Rhinolophus (bat), Ursus
(bear), Castanea (chestnut), Corylus (hazelnut), and Linum (ax) (Fig.5 and SI Table3). e combined number
of uniquely aligned reads to the reference species of these genera (SI Tables3 and 4) per speleothem section varied
between 4541 (SKK12) and 72056 (SKK3), and the per-species damage patterns between 0 and 54% (Fig.5, SI
Figure 4. e cut stalagmite SKK 16 5 (photo taken by MCS). SKK 16 was sampled for U-series dating (red dotted
line, unsuccessful analyses) and DNA analysis (green, dotted line if unsuccessful analysis), which include samples
from the stalagmites as well as the incorporated bones (SKK 7 and 12 from cortical bone and SKK 10 from trabecular
bone). Similar to SKK 16 3, stalagmite SKK 16 5 was also partially cut open with a rock saw and then broken open
(le of the blue dashed line) and both contexts were sampled. Note the brown colour of the speleothem, indicating the
presence of detrital clay, which impeded the U-series dating. However, each sample gave aDNA reads.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

6
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
Table3). e genus most frequently found was Capreolus (roe deer) which was positively identied in 7 of the 11
samples, followed by Rhinolophus (bat) in six, Castanea (chestnut) in ve, Ursus (bear) and Corylus (hazelnut)
in four, and Linum (ax) in only two (Fig.5, SI Table3). In SKK 10 we conrmed the presence of six ancient
genera, the highest number among all Solkota samples. Interestingly, the samples from the bear bone embedded
within the speleothems (SKK7, 10 and 12) also contained exogenous aDNA, including aDNA of other mammalia
genera and plantae with damage patterns ranging from 11.87 to 27.50% (Fig.5). e samples from the bear bone
embedded in the speleothem matrix provided high numbers of aligned reads to Ursus (8465 for SKK7, but only
126/125 for SKK 10/12) and display a strong deamination pattern above 50% for SKK 7 and nearly 30% for SKK
10 and 12 (Fig.5, SI Table3). Together, the aligned reads for bear, the clear damage pattern, the minimum age of
the bone and the zooarchaeological observations indicate that the speleothem embedded bone originates from
cave bear. Sample SKK 1 also displayed a strong deamination pattern for bear reads, nearly 50%, here, however, no
bear bone was present. Negative control analysis identied no ancient molecules aligned to any of the mentioned
genomes (SI Table3), indicating no cross-contamination between samples.
Discussion and Conclusion
Growth phases of stalagmite SKK 16 3 can be linked to global climatic records. e speleothem’s rapid but inter-
mittent growth around 84 ka coincides with climatic amelioration during Marine Isotope Stage (MIS) 5a, the
Greenland Interstadial 21.1e (GI-21.1e)37. e resumption of growth at 56.7 ± 0.95 coincides with a warm interval
in MIS3, Interstadial GS 16.137. For both time periods, interstadial GI-21e and GI 16.1, no dates for human occu-
pation in the region have been reported. However, this may be the result of limited dating of human occupation
deposits beyond the range of radiocarbon in this region, many Middle Paleolithic sites still lack absolute dating
(Bronze Cave, Sakaja and Ortvala38, Koudaro I, Undo39, Djruchula and Tsona40). Speleothem growth at Solkota
Cave suggests episodic favourable climatic condition in the region during parts of MIS5 and MIS3, which could
also have supported human occupation. However, climatic interpretations need to be further investigated with
stable isotope data and can now also be coupled with environmental aDNA from the same stalagmite.
Our rst metagenomic analyses presented here allowed the documentation of aDNA from inside the stalag-
mites with characteristic deamination damage to the DNA. We were able to identify the aDNA inside the stalag-
mites down to genera and to show the preservation of aDNA from mammals (bear, roe deer, horseshoe bat) and
plants (chestnut, hazelnut, ax) from various layers inside the speleothem as well as from the incorporated bone
(Fig.6). e identied plants and large mammals indicate a generally forested environment. Similar landscape is
also reconstructed from later Paleolithic sites of the area, such as Kotias Klde41,42 and Satsurblia33,42. Apart from
sample SKK 2, all samples from stalagmites SKK 16 3 and 5 contained aDNA from one or more genera. SKK 1 and
3 from stalagmite SKK 16 3 gave each one genera conrmation, bear and roe deer respectively. For stalagmite SKK
16 5, the number of detected genera range from 1 to 6 per sample. Bone samples from this stalagmite (SKK7, 10,
12) exhibit a higher number of conrmed genera (4–6 per sample) than pure speleothem samples (SKK4, 5, 6, 8,
9, 11) (1–3 per sample). e preservation of aDNA with characteristic deamination damage in most of the studied
samples show that both stalagmite and bone embedded in stalagmites are a promising medium for aDNA preser-
vation. However, only for the aDNA from stalagmite SKK 16 3 absolutes ages could be inferred. SKK 3, containing
roe deer, can be condently assigned to be older than ~84 ka. e same layer that preserved the ancient bear DNA
in SKK1, between hiatus 3 and the combined hiatus 1 and 2, was dated to 83.32 ± 1.48 ka with U-series. However,
this layer is rather thin and it is possible that the aDNA sample contained dust from either hiatus event and the
bracketing the U-series dates, 84.57 ± 0.76 ka (older layer) and 56.7 ± 0.95 ka (younger layer), provide a more
reliable chronological frame. For stalagmite SKK 16 5 only a minimum age for the bone, >50.200 uncalibrated,
could be deduced and the age of the stalagmite, which formed aer the deposition of the bone, remains open.
The diverse occurrence of the aDNA with variability inside the stalagmites, between sampling contexts
(pure speleothem versus bone embedded in the speleothem), between stalagmites from the same cave and from
Figure 5. Deamination frequencies in the Solkota samples (graph made by DMF). e graph presents the
average deamination frequencies at the 5′ and 3′ bases for the terminals ends. Only genera exceeding a 10%
deamination threshold were accepted as ancient and are presented here. Solid bars represent mammalian
genera, patterned bars represent plantae.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

7
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
dierent caves of the same cave system, Solkota and Melouri Cave, open up the question of the formation history
of the aDNA in this context. is formation history includes the DNA source, DNA adsorption, transport (agent),
deposition and preservation of DNA inside the stalagmites. Zepeda Mendoza et al.32 noted in their analysis,
that aDNA inside the studied popcorn speleothem contained aDNA originating from outside the cave as well as
from dierent parts inside the cave. Similarly, in our study aDNA from cave dwelling genera (bear, bat) as well as
non-cave dwelling genera (roe deer, hazelnut, chestnut and ax) are present. is mixture of allochthonous and
autochthonous sources suggests also a mixture of depositional processes. A number of possible processes can be
imagined. First, water is one possible transport agent, which inltrates through soils above the cave through the
epi-karst system into the cave, transporting plant, animal, bacterial, insect and fungal DNA. Another possible
biogenic process is the direct contact of the organism, from which the DNA derived, with the speleothem, e.g.
bears rubbing on speleothems, food remains (roe deer, nuts) adhering to the bear and being transported into
the cave, bats and bear defecating and urinating. A nal possible process is the gravitational transport of DNA
adhering to sediment particles into the cave, as can easily be imagined with the steep entrance slope at Solkota
cave. Clays, ne organic matter, silica grains and other minerals can all occur inside speleothems32,43,44. Aer
deposition and adsorption of the DNA and formation of the speleothem, DNA preservation and integrity is pro-
moted by the closed system of the stalagmite, making them a probably more reliable archive than sedimentary
deposits and soils. In addition, precise dating of layers containing aDNA is possible. However, possible minor
post-depositional migration of uranium in the lower part of speleothem SKK 16 5 (the lowermost age is out of
stratigraphic order) opens up the question of the integrity of speleothem to post-depositional migration of aDNA.
is being said, if DNA fragments are strongly adsorbed to sediment components, e.g. clay minerals17–19 depos-
ited during the hiatus events within the speleothem, post-depositional migration by drip waters that percolate
through the speleothem structure is less likely for the DNA than for water-soluble uranium.
e preservation of aDNA inside speleothems entails diverse prospects for archaeological and paleoenviron-
mental research. Paleoenvironmental speleothem records from cave sites are associated with contemporaneous
archaeological, paleontological and paleobotanical records via correlating dates. e detection of mammalian and
plant aDNA inside speleothems reveals a potentially direct link between these records. Furthermore, stalagmites
can serve as an additional archive for old excavation, where all sediments, archaeological and paleontological
remains have already been removed, or for sites where bone and overall organic preservation is poor.
Materials and Methods
For sample extraction and to limit/control for sample contamination, speleothems were sawed only partially open
with a rock saw, using deionized water for cooling, and were then broken open to reveal surfaces for sampling.
Aer stratigraphic interpretation of the speleothems, samples were taken with a micro drill using layer-parallel
elliptical sampling pits and including samples from the sawed and broken area (Figs3 and 4). U-series samples
Figure 6. Genera identied by the aDNA analyses in stalagmites SKK 16 3 and 5 (photos taken by MCS). ©
MPI for Evolutionary Anthropology.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

8
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
were taken to constrain the age of calcite layers that contain aDNA and some were taken in tandem with the DNA
samples, from adjacent/overlapping sampling locations. Sample size was 100–200 mg for U-series and 25–60 mg
for DNA analysis. Samples for the latter were taken in a clean laboratory, using bleach to clean the surface of the
speleothems and about 2 mm of the exposed inner speleothems surface was removed with the micro drill before
sampling to limited contamination. Samples for DNA sequencing include samples from the speleothem as well as
samples from the bone inside speleothem SKK 16 5, cortical and trabecular bone (Fig.4).
U-series. Methods for U-series dating methods were similar to those described by Fankhauser et al.44.
Briey, sample powders were weighed and spiked with a mixed 233U/236U/229 tracer. Following dissolution and
spike equilibration, separation of uranium and thorium was completed by anion exchange chromatography. All
measurements were carried out using a ermoFisher Neptune® high-resolution inductively coupled plasma
mass spectrometer with an Aridus® desolvation nebuliser at the School of Earth Sciences, University College
Dublin. 238U/236U and 233U/236U ratios were measured using three Faraday collectors, while the 234U ion beam
was measured in a secondary electron multiplier (SEM). Calibration of the SEM relative to the Faraday detectors
was achieved by sample-standard bracketing, using the certied 235U/238U ratio of the IRMM-3184 standard.
Mass-fractionation corrections for uranium were applied based on the certied 233U/236U ratio of the mixed spike.
e minor isotopes of thorium (230 and 229) were measured using the SEM, whilst two Faraday collectors
were used to simultaneously measure the much larger 232 ion beam. A standard-sample bracketing method
using the IRMM-318444 standard a uranium standard was applied for the thorium mass fractionation correction
and for the SEM/Faraday yield calibration.
As discussed in the results, several sub-samples from the speleothem contained significant amounts of
non-carbonate ‘detrital’ thorium as evidenced by high 232 concentrations and low 230/232 ratios (Table1).
is necessitated corrections for inherited non-radiogenic 230. In the literature this correction is oen achieved
by simply assuming that the 230/232 ratio of the inherited (non-carbonate) fraction is equivalent to a typical
upper crustal 238U/232 activity ratio of 0.8 ± 0.445. In this study, in order to reduce the dating uncertainties asso-
ciated with the detrital correction we measured the U-series isotope ratios in clay-rich samples from the cave to
constrain the actual non-carbonate (silicate clay mineral) 230/232 ratio.
Ancient DNA. DNA samples were extracted and prepared within a clean room environment at a dedicated
ancient DNA laboratory at University College Dublin (UCD), Ireland. Unilateral air-ow hoods, tyvek suits,
hair nets, face masks and non-powdered gloves were used to limit contamination. Upon amplication further
steps were performed in a modern laboratory environment. DNA extraction was undertaken according to the
method outlined by Collin et al.46,47 (Collin manuscript in preparation). is protocol, developed for the extrac-
tion of aDNA from anthropogenic sediments, reduces the action of potentially damaging geopolymers on DNA
by chemical inhibition and increases the range as well as quantity of isolated DNA fragments thereby reducing
dependency on DNA capture techniques for exploratory samples. e extraction protocol consisted of samples
being placed into Matrix E lysing tubes (MP-BIO-116914050) and submerging them in 1 mL of extraction buer
up to a nal concentration of 0.45 M EDTA, 0.02 M TrisHCL (pH 8.0), 0.025% SDS, 0.5 mg/mL Proteinase K and
dH2O. Samples were incubated at 39 °C overnight using an Eppendorf ermomixer® C with a rotational speed
of 1600rpm to ensure maximal bead movement. Supernatant was collected and cleaned following Dabney et al.48
and DNA libraries were prepared following Meyer and Kircher49. Negative controls were included at all stages and
pooled to investigate the presence of damaged reads indicative of cross-contamination during DNA extraction
and library preparation.
DNA samples were amplied using a universal Illumina primer and Polymerase chain reaction (PCR) follow-
ing Gamba et al.50 and were repeated 15 times following Collin et al.46,47. Assessment of PCR reaction concen-
trations were performed on an Agilent 2100 Bioanalyser following instructions of the manufacturer. Based on
these concentrations samples were pooled into a 4 nM working solution and sequenced on an Illumina NextSeq.
500/550 using the high output v2 (75 cycle) reagent kit at UCD Conway Institute of Biomolecular and Biomedical
Research. Genera were initially identied by cross referencing raw sequencing data with the National Centre for
Biotechnology (NCBI) genomic database using Basic Local Alignment Search Tool (BLAST)51–53 at an evalue of
1e-05 and MGmapper54 with a 0.8 fraction of matches + mismatches and minimum alignment score of 2547. In
order to reduce the chances of false positive identications of taxa at the family or genus level we followed a sim-
ilar approach to that of Slon et al.15. Aer the initial alignment using Blastn, an oine nucleotide BLAST + data-
base was generated with the genomic sequences of the 16 main eukaryote species detected (12 animals and 4
plants, SI Table2) with “makeblastdb” (genome versions for each species presented in SI Table4). Each sample’s
trimmed reads were aligned to this database using default “blastn” parameters and the resulting output data
was imported into MEGAN Community Edition v.6.2.1355. For the last common ancestor (LCA) parameters
we used a minimum bitscore of 35 within the top 10% of the best alignments, minimum support count of 2, and
the default “naive” LCA algorithm15. A minimum 1% of the total assigned reads was necessary to accept a taxa
to be present following Slon et al.15 and were then used for downstream analysis (SI Table3). e sets of reads
assigned to each species were extracted into independent les and then aligned to the correspondent genome
for authentication. We used BWA v.0.7.5a-r405 “aln”56 with permissive parameters (−o 2 −n 0.01) and disabled
seed (-l 1000), and then the aligned reads were ltered for a minimum quality of 25, sorted, duplicates removed,
and indexed using samtools v.1.3.157. Using mapDamage257 we investigated and quantied the presence of C to
T substitutions on the 5’ end and G to A on the 3’ end of the sequences, and used a minimum value of 10% on
both sides for a taxon to be identied as ancient6,16. Average read lengths were calculated using Genome Analysis
Toolkit’s “ReadLengthDistribution” (see SI Fig.3 for an averaged deamination length plot)58.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
References
1. Orlando, L. et al. ecalibrating Equus evolution using the genome sequence of an early Middle Pleistocene horse. Nature 499, 74–78
(2013).
2. Shapiro, B. & Hofreiter, M. A Paleogenomic Perspective on Evolution and Gene Function: New Insights from Ancient DNA. Science
343, 1236573 (2014).
3. Barlow, A. et al. Partial genomic survival of cave bears in living brown bears. Nat. Ecol. Evol. 2, 1563–1570 (2018).
4. rause, J. et al. A complete mtDNA genome of an early modern human from osteni, ussia. Curr. Biol. 20, 231–236 (2010).
5. rause, J. et al. e complete mitochondrial DNA genome of an unnown hominin from southern Siberia. Nature 464, 894–897
(2010).
6. Prüfer, . et al. e complete genome sequence of a Neanderthal from the Altai Mountains. Nature 505, 43–49 (2014).
7. Jones, E. . et al. Upper Palaeolithic genomes reveal deep roots of modern Eurasians. Nat. Commun. 6, 8912 (2015).
8. Posth, C. et al. Pleistocene mitochondrial genomes suggest a single major dispersal of non-Africans and a Late Glacial population
turnover in Europe. Curr. Biol. 26, 827–833 (2016).
9. Wang, C.-C. et al. e genetic prehistory of the Greater Caucasus. bioxiv 322347, https://doi.org/10.1101/322347 (2018).
10. Hofreiter, M., Mead, J. I., Martin, P. & Poinar, H. N. Molecular caving. Curr. Biol. 13, 693–695 (2003).
11. Parducci, L. et al. Ancient plant DNA in lae sediments. New Phytol. 214, 924–942 (2017).
12. Murray, D. C. et al. High-throughput sequencing of ancient plant and mammal DNA preserved in herbivore middens. Quat. Sci. ev.
58, 135–145 (2012).
13. Willerslev, E. et al. Fiy thousand years of Arctic vegetation and megafaunal diet. Nature 506, 47–51 (2014).
14. Haile, J. et al. Ancient DNA reveals late survival of mammoth and horse in interior Alasa. Proc. Natl. Acad. Sci. 106, 22352–22357
(2009).
15. Slon, V. et al . Neandertal and Denisovan DNA from Pleistocene sediments. Science 356, 605–608 (2017).
16. Briggs, A. W. et al. Patterns of damage in genomic DNA sequences from a Neandertal. Proc. Natl. Acad. Sci. 104, 14616–14621
(2007).
17. Goring, C. A. I. & Bartholomew, W. V. Adsorption of mononucleotides, nucleic acids, and nucleoproteins by clays. Soil Sci. 74,
149–164 (1952).
18. Greaves, M. P. & Wilson, M. J. e adsorption of nucleic acids by montmorillonite. Soil Biol. Biochem. 1, 317–323 (1969).
19. Cai, P., Huang, Q., Zhang, X. & Chen, H. Adsorption of DNA on clay minerals and various colloidal particles from an Alsol. Soil
Biol. Biochem. 38, 471–476 (2006).
20. Lorenz, M. G. & Wacernagel, W. Adsorption of DNA to sand and variable degradation rates of adsorbed DNA. Appl. Environ.
Microbiol. 53, 2948–2952 (1987).
21. Bezanilla, M., Manne, S., Laney, D. E., Lyubcheno, Y. L. & Hansma, H. G. Adsorption of DNA to mica, silylated mica, and minerals:
characterization by atomic force microscopy. Langmuir 11, 655–659 (1995).
22. Crecchio, C. & Stotzy, G. Binding of DNA on humic acids: eect on transformation of Bacillus subtilis and resistance to DNase. Soil
Biol. Biochem. 30, 1061–1067 (1998).
23. Cleaves, H. J. II et al. e adsorption of short single-stranded DNA oligomers to mineral surfaces. Chemosphere 83, 1560–1567
(2011).
24. Haile, J. et al. Ancient DNA chronology within sediment deposits: Are paleobiological reconstructions possible and is DNA leaching
a factor? Mol. Biol. Evol. 24, 982–989 (2007).
25. awlence, N. J. et al. Using palaeoenvironmental DNA to reconstruct past environments: progress and prospects. J. Quat. Sci. 29,
610–626 (2014).
26. McDermott, F. Palaeo-climate reconstruction from stable isotope variations in speleothems: a review. Quat. Sci. ev. 23, 901–918
(2004).
27. Cadet, J. & Wagner, J. . DNA base damage by reactive oxygen species, oxidizing agents, and UV radiation. Cold Spring Harb.
Perspect. Biol. 5, a012559 (2013).
28. astogi, . P., icha, umar, A., Tyagi, M. B. & Sinha, . P. Molecular mechanisms of ultraviolet radiation-induced DNA damage
and repair. J. Nucleic Acids 2010, 592980 (2010).
29. Spear, J. ., Barton, H. A., obertson, C. E., Francis, C. A. & Pace, N.  . Microbial community biofabrics in a geothermal mine adit.
Appl. Environ. Microbiol. 73, 6172–6180 (2007).
30. Yücel, S. & Yamaç, M. Selection of Streptomyces isolates from Turish arstic caves against antibiotic resistant microorganisms. Pa
J Pharm Sci 23, 1–6 (2010).
31. Or tiz, M. et al. Maing a living while starving in the dar: metagenomic insights into the energy dynamics of a carbonate cave. ISME
J. 8, 478–491 (2014).
32. Zepeda Mendoza, M. L. et al. Metagenomic analysis from the interior of a speleothem in Tjuv-Ante’s Cave, Northern Sweden. PloS
One 11, e0151577 (2016).
33. Pinhasi, . et al. Satsurblia: new insights of human response and survival across the Last Glacial Maximum in the southern
Caucasus. PloS One 9, e111271 (2014).
34. Bar-Yosef, O. et al. Dzudzuana: an Upper Palaeolithic cave site in the Caucasus foothills (Georgia). Antiquity 85, 331–349 (2011).
35. Meshveliani, T. et al. Mesolithic hunters at otias lde, western Georgia: preliminary results. Paléorient 33, 47–58 (2007).
36. uuri, T. et al. Geomorphologic Features of Sataplia-Tsaltubo Limestone Massif. Earth Sci. Spec. Issue Mod. Probl. Geogr.
Anthropol. 4, 5–1 (2015).
37. asmussen, S. O. et al. A stratigraphic framewor for abrupt climatic changes during the Last Glacial period based on three
synchronized Greenland ice-core records: rening and extending the INTIMATE event stratigraphy. Quat. Sci. ev. 106, 14–28
(2014).
38. Pinhasi, . et al. New chronology for the Middle Palaeolithic of the southern Caucasus suggests early demise of Neanderthals in t his
region. J. Hum. Evol. 63, 770–780 (2012).
39. Moncel, M.-H. et al. The Middle Palaeolithic ecord of Georgia: A Synthesis of the Technological, Economic and
Paleoanthropological Aspects. Anthropologie 53, 93–125 (2015).
40. Tushabramishvili, N., Pleurdeau, D., Moncel, M.-H. & Mgeladze, A. Le complexe Djruchula-oudaro au sud Caucase (Géorgie).
emarques sur les assemblages lithiques pléistocènes de oudaro I, Tsona et Djruchula. Anthropologie XLV, 1–18 (2007).
41. Bar-Oz, G. et al. Bear in Mind: Bear Hunting in the Mesolithic of the Southern Caucasus. Archaeol. Ethnol. Anthropol. Eurasia. 37,
15–24 (2009).
42. Martoplishvili, I. & vavadze, E. Some popular medicinal plants and diseases of the Upper Palaeolithic in Western Georgia. J.
Ethnopharmacol. 166, 42–52 (2015).
43. Blyth, A. J., Hartland, A. & Baer, A. Organic proxies in speleothems – New developments, advantages and limitations. Quat. Sci.
ev. 149, 1–17 (2016).
44. Fanhauser, A., McDermott, F. & Fleitmann, D. Episodic speleothem deposition tracs the terrestrial impact of millennial-scale last
glacial climate variability in SW Ireland. Quat. Sci. ev. 152, 104–117 (2016).
45. Homann, D. L. et al. U- dating of carbonate crusts reveals Neandertal origin of Iberian cave art. Science 359, 912–915 (2018).
46. Collin, T. C., Pinhasi, . & Feeney, . M. N. Optimisation of metagenomic next generation sequencing shotgun techniques for the
study of ancient anthropogenic sediments. Am. J. Phys. Anthropol. Suppl. S62, 119 (2016).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

10
SCIENTIFIC REPORTS | (2019) 9:6628 | https://doi.org/10.1038/s41598-019-43147-0
www.nature.com/scientificreports
www.nature.com/scientificreports/
47. Collin, T. C., Stahlschmidt, M. C., Pinhasi, . & Feeney, . M. N. Metagenomic study of anthropogenic sediments: insights into
public health and lifestyle. Human Evolution: Fossils, Ancient and Modern Genomes, 20–22 November 2017, Wellcome Genome
Campus, Hinxton, Cambridge, U (2017).
48. Dabney, J. et al. Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA
fragments. Proc. Natl. Acad. Sci. 110, 15758–15763 (2013).
49. Meyer, M. & ircher, M. Illumina sequencing library preparation for highly multiplexed target capture and sequencing. Cold Spring
Harb. Protoc. 2010, pdb–prot5448 (2010).
50. Gamba, C. et al. Genome ux and stasis in a ve millennium transect of European prehistor y. Nat. Commun. 5, 5257 (2014).
51. Altschul, S. F. & Gish, W. Local alignment statistics. In Methods in Enzymology 266, 460–480 (Academic Press 1996).
52. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local alignment search tool. J. Mol. Biol. 215, 403–410 (1990).
53. Altschul, S. F. et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids es. 25,
3389–3402 (1997).
54. Petersen, T. N. et al. MGmapper: eference based mapping and taxonomy annotation of metagenomics sequence reads. PloS One
12, e0176469 (2017).
55. Huson, D. H. et al. MEGAN Community Edition - Interactive Exploration and Analysis of Large-Scale Microbiome Sequencing
Data. PLOS Comput. Biol. 12, e1004957 (2016).
56. Li, H. et al. e Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
57. Jónsson, H., Ginolhac, A., Schubert, M., Johnson, P. L. F. & Orlando, L. mapDamage2.0: fast approximate Bayesian estimates of
ancient DNA damage parameters. Bioinformatics 29, 1682–1684 (2013).
58. Mcenna, A. et al. e Genome Analysis Toolit: a Mapeduce framewor for analyzing next-generation DNA sequencing data.
Genome es. 20, 1297–1303 (2010).
59. NASA/METI/AIST/Japan Spacesystems & .S./Japan ASTE Science Team. ASTE Global Digital Elevation Model. NASA EOSDIS
Land Process. DAAC, https://doi.org/10.5067/ASTE/ASTGTM.002 (2009).
Acknowledgements
e research presented here was supported by a New Interdisciplinary Initiatives Fund grant (SF1362) to MCS
by the University College Dublin. Further support came from a Government of Ireland Postdoctoral Fellowship
to MCS (GOIPD/2015/775), a Medical Trainee PhD Scholarship, Anatomy, School of Medicine, University
College Dublin awarded to TCC and from the Moshe and Bina Stekelis Foundation and Moshe Stekelis Chair in
Prehistoric Archaeology for excavation at Satsurblia Cave in 2016 to ABC, hosting our eld expedition. We are
grateful to the excavation team of Satsurblia, the people of Kumistavi and the Georgian State Museum. We further
would like to thank Mick Murphy from the UCD School of Earth Sciences for assistance with the U-series lab
work and Susanna Sawyer from Department of Evolutionary Anthropology, University of Vienna, for assistance
with sequencing data interpretation. Figure1 was created with ASTER GDEM, a product of METI and NASA,
and with help from R. C. Power, Max-Planck Institute of Evolutionary Anthropology. We are grateful to S. Tüpke
and the Max-Planck Institute of Evolutionary Anthropology for making the silhouettes in Fig.6.
Author Contributions
M.C.S. and R.P. conceived and supervised the project with contributions from T.C.C. and F.M.D. M.C.S. wrote
the paper with contributions from R.P., T.C.C., F.M.D., G.B.O. and D.M.F. M.C.S. and F.M.D. performed the
u-series dating, T.C.C., D.M.F., Z.G., J.P., R.P. the DNA analysis. A.B.C. and G.B.O. identied lithics and bones
respectively. M.C.S., F.M.D., R.P., A.B.C., G.B.O., N.J., Z.M. and T.M. conducted eld work related to the project.
Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-43147-0.
Competing Interests: e authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2019
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com