Samples of tissues of mummies under the names "Maria" (Maria) and "Vavita" (Wawita) were sent from Peru to a Russian laboratory for research. The provided samples of biological tissues were studied using scanning electron microscopy, Raman spectra, ICPE, and the composition of diatomaceous earth (a substance on the surface of the skin of mummies) was investigated. A study of the DNA of the transferred tissues was also carried out.
Sample preparation
Tissue samples of 500 μg were transferred into plastic tubes and dissolved in 1 ml of buffer (10 mM Tris-HCl pH = 10.5, 1 mM EDTA, 0.15 M NaCl). To the resulting suspensions, Na dodecyl sulfate was added to 0.5%, heated to +80 C, and after 10 minutes, proteinase K (up to 500 μg / ml) was placed in a thermostat (+55 C) for 24 hours.
Deproteinization was carried out by the phenolic method: adding phenol in equal volume to the suspension, then phenol: chloroform (1: 1), chloroform; after each addition, constant stirring was carried out on an angular rotor and centrifugation at 15 thousand rpm, 10 min.
To the super-sediment obtained after the 3rd centrifugation was added 1/10 of the volume of 1 M NaCl and 2.5 volumes of twice distilled ethyl alcohol, and left overnight at -30 C in conical test tubes - "Eppendorf". Centrifugation was carried out at 15 thousand vol. min. within 10 minutes and received DNA "precipitated" in the form of a brown precipitate. The DNA pellet was washed twice with 70% ethyl alcohol and dried at room temperature (1 h) and then dissolved in TE buffer.
PCR was performed on a programmable thermal cycler "My Cycler" ("Bio = Rad") using standard oligoprimers synthesized by the solid-phase method at the association "Beagler" (St. Petersburg). The reaction mixture for amplification with a volume of 25 μL included: 15 nM of each oligoprimer, 67 mM Tris-HCI pH = 8.8 at +25 C, 16.6 mM (NH4) SO4, 6.7 mM MgCl2, 6.7 μM EDTA, 10 mM 2 mercaptoethanol, 170 μg / ml BSA and a mixture of four basic dNTPs at a concentration of 1.0 mM each and thermostable DNA polymerase Thermus thermophilis (5 U / μl) (NPO SibEnzim). After denaturation (10 min, 94 C), 35 amplification cycles were performed for each test system: 94 ° C -1 min, 58 ° C - 1 min, 72 ° C - 1 min. To control the specificity, DNA samples with known genotypes for the studied loci (marker systems), as well as control samples, were introduced into the reaction.containing a mixture of reagents without DNA. After amplification, a staining buffer was added to aliquots of the reaction mixture (7 μl) and separated by vertical electrophoresis in 6% polyacrylamide gel (210x150x1 mm), stained with ethidium bromide and photographed under ultraviolet light. To identify alleles, allelic standards corresponding to these loci ("ladders") were used.
Promotional video:
DNA analysis
For the study, we used the method of exome analysis of human DNA based on high-throughput sequencing with enrichment by hybridization.
Technique of exome analysis of human DNA.
2 samples were analyzed … All stages of sample preparation before PCR were carried out in clean rooms. Sample preparation, DNA extraction, and amplification of individual DNA fragments were performed in different rooms.
Specific adapters (KAPA Library Preparation Kit and SeqCap Adapter Kit; Roche) were ligated to the ends of the genomic fragmented DNA (~ 5 ng), after which a two-stage selection of fragments in the 200-350bp length range was carried out using AMPureXP Beads (Beckman Coulter) … The resulting fragments were amplified with adapter specific primers and hybridized with biotinylated specific probes (NimbleGen SeqCap EZ Choice MedExome; Roche) for 28 h at 47 ° C. In one reaction, two DNA samples were combined. Biotinylated probe-DNA hybrids were isolated and purified with streptovidin conjugated magnetic particles; a second amplification and qualitative assessment of the resulting DNA library was performed (TapeStation 4200; Agilent Technologies). In order to remove off-target amplification fragments and adapter dimers, the DNA library was re-purified using AMPureXP Beads magnetic particles (Fig. 1). The final concentration of the prepared library was evaluated on a Quantus instrument using a commercial QuantiFluor® dsDNA System kit (Promega). The resulting DNA library was immobilized onto the surface of the flow cell. Sequencing was performed on the Illumina platform using a Standard Flow Cell and MiSeq Reagent Kit v2 300 (2 × 150 cycles). The resulting DNA library was immobilized onto the surface of the flow cell. Sequencing was performed on the Illumina platform using a Standard Flow Cell and MiSeq Reagent Kit v2 300 (2 × 150 cycles). The resulting DNA library was immobilized onto the surface of the flow cell. Sequencing was performed on the Illumina platform using a Standard Flow Cell and MiSeq Reagent Kit v2 300 (2 × 150 cycles).
Figure: 1
General assessment of the quality of sequenced DNA
Standard methods such as FastQC and the Jellyfish and KraTER programs were used for the initial quality assessment. The quality assessment showed no quality problems with the sequenced DNA reads for both samples. Pictures are not shown as they are not informative.
For the first sample (further M, large mummy "Maria") 113.4M reads were sequenced, for the second sample (further W, small mummy "WaWita") 22.9M reads were sequenced.
The next step was to cleanse the sequenced reads from various technical sequences that would interfere with further analysis. For this, the Cookiecutter program was used. After cleaning, there were 113.3M (99%) reads and 22.8M (99%) reads, respectively, for M and W.
Estimation of the amount of ancient DNA and its separation from modern
The standard method for assessing the presence and amount of ancient DNA is to estimate the amount of time-damaged nucleotides. For this, the MapDamage 2.0 program was used. MapDamage 2.0 which showed the amount of ancient DNA in 30.1%, but since MapDamage implies that we are using a close reference, and we do not know exactly how close both samples are to the genome of modern humans, and we used an exome library, this method itself was not sufficient. Another good method for assessing ancient DNA is the number of short fragments.
Filtration of the alleged ancient DNA took place in three stages. At the first stage, we removed all paired reads that cannot be crossed and assembled into one unpaired read with an overlap. These reads indicated that the original fragment that was sequenced was longer than 300 bp. and most likely modern DNA. So after this step, 86.9% and 91.8% remained, respectively. After that, single overlapping fragments were selected so that their length was shorter than 150 bp, since this is the length we expected for ancient DNA. After this step, 8.6% and 38.5% remained, respectively, for samples M and W. It should be noted that despite the difference in percentage, the absolute number of reads between samples M and W is very similar: 9.8M and 8.8M, which can be explained by the fact that the content of ancient DNA in both samples is similar.
| Parameter | Sample M (Maria) | Sample W (Wawita) |
| Number of reads | 113.3M | 22.8M |
| Percentage of short fragments <300 bp | 86.9% | 91.8% |
|
Percentage of short fragments <150 bp (number) |
8.6% (9,846,035) |
38.5% (8 813 220) |
|
Percentage of fragments aligned per human genome (number) |
2.03% (2,345,084) |
9.65% (2,264,551) |
| Percentage of fragments aligned per human genome from short <150 bp | 23.8% | 25.6% |
Obtaining DNA similar to the reference human genome
The resulting putative ancient DNA was mapped to a reference human genome using a standard genomic variant search pipeline using BWA, samtools, and Wcftools.
Moreover, only 23.8% of sample M and 25.6% were successfully mapped to the reference genome of modern humans. At the same time, for both samples, 75% of the sequenced reads could not be mapped onto the human genome. This can be explained both by contamination and by the fact that these samples are located far enough from the genome of modern humans. At the same time, it is worth keeping in mind that we have sequenced the exome library and thereby minimized the contamination of bacterial DNA.
Initial reconnaissance analysis of uncooked reads showed that some of them belonged to repetitive DNA specific to ungulates, this can be explained by the fact that llama fat was used in mummification.
For a more detailed analysis, it takes about three weeks of calculations, since the existing solutions were made only for viral and bacterial genomes, and we need to compare with all existing genomes, including plant genomes, in order to understand which species were sequenced.
Characteristics of the found options
The mapped reads were used to search for variants that distinguish M and W samples from the modern human genome, as well as to assess contamination of reads from the human Y chromosome.
The first question that needed to be answered was which chromosomes the sequenced reads were mapped to.
Since we know that Maria's samples were isolated from bones and muscles, and Vavita's only from bones, we expected a difference in the amount of mtDNA. But there wasn't much of a difference.
The number of mapped reads on Y is another test for contamination by modern humans. Interestingly, it turned out to be the same in both samples.
The statistics for the variants found are shown below:
| Parameters | Sample M (Maria) | Sample W (Wawita) |
| Number of options found | 79957 | 48941 |
| Number of options on Y | 534 | 541 |
| The number of reliable options (more than 20 reads and more than 30 quality) | 16969 | 6181 |
| Variants with known rsid (available in the snip database) | 5701 | 3089 |
| General valid options | 92 | |
| common options with rsid | 49 |
After that, it became possible to answer the question: are M and W relatives? The answer is no.
Only 49 matching variants were found between M and W and 3040 differing for variants with known rsid. Moreover, perhaps these are different types or subspecies of a person or an unknown creature.
Interestingly, the variants with the Y chromosome are identical for both samples, which indicates contamination by the same person, and that in the ancient DNA of the Y chromosome, there is probably no chromosome.
Evaluation of the similarity with the existing sequenced human genomes from the 1000 Human Genome Project
Note that this is only an approximate analysis, since for a more accurate analysis, variants are needed from regions of the genome under neutral selection, and we have exome data.
Nevertheless, using 5708 variants for M or 3096 for S, it was possible to carry out a variant of the analysis compared with the data of 1000 human genomes.
The result of the PCA analysis in the picture below is an overlay of two pictures for M and W, calculated separately, since there are too few common options between M and W to estimate the distances between M and W.
Similarity score.
As you can see, there is no coincidence with any group of genes, they also differ from each other. But it should be borne in mind that we used coding sequences under selection, and it is recommended to use variants under neutral selection.
Nevertheless, the PCA result is in good agreement with manual verification of variants, which showed that the data are in an unreferenced homozygote, which again indicates that the images are far from the modern human genome.
Conclusion
Unfortunately, we were limited to only two samples, usually in this type of analysis more are used, at least 3-10 at least somehow related. Therefore, it is necessary to continue research with a large number of samples.
At the same time, we can most likely conclude that the DNA samples of Mary and Vavita correspond to human DNA, but do not coincide with the DNA available to us from a database of 1000 people.
Authors of the report: Baranov V. S. and Aseev M. V. (Scientific Research Institute of Obstetrics and Gynecology, Department of Prenatal Diagnostics), Glotov A. S. and Glotov O. S. (St. Petersburg State University), A. S. Komissarov (Institute of Cytology, Russian Academy of Sciences, Center for Genetic Bioinformatics).
Materials provided by Konstantin Georgievich Korotkov (Doctor of Technical Sciences, Professor, University of Information Technologies, Mechanics and Optics) and Dmitry Vladislavovich Galetsky (Candidate of Medical Sciences, I. P. Pavlov First St. Petersburg State Medical University).
For more about Nazca mummies, see the tag: Nazca mummies.
