Ancient microbiomes/pathogens & forensic genetics Flashcards
(28 cards)
What is the difference between microorganisms, microbiota and microbiome?
Microorganisms: microscopic organisms living in all kinds of environments
Microbiota: a community of microorganisms in a specific environment
Microbiome: a community of microorganisms and its environmental and genomic context that lives together in a certain location or niche.
Microbiomes are everywhere! Give three examples.
- Animals have microbiomes, for example for digesting certain kinds of diet.
- Plants have specific microbiomes on their above soil part and in their root system
- Humans generally have four distinct microbiomes: Skin, Mouth, Gut and urogenital tract (but for example lungs also).
What types of microorganisms are there?
- Bacteria (bacteriome): most abundant by far.
- Fungi (mycobiome): oral, gut, skin and lungs.
- Viruses (virome): e.g. HPV, EBV, HHV-6B (Herpesvirus 6) etc. Many stay and incorporate into the genome after you “pass” them.
- Archaea (archaeome). frequently overlooked :(
- Protozoa (single-celled eucaryotes parasitic or not): high prevalence in healthy populations but pathogenic way more studied.
- Algae (not in humans though)
Pathogenic microorganisms are generally more studied than commensals (unfortunately - but priority makes sense)
How much microbial cells in relation to human cells are there in the human body?
The ratio of human to microbial cells are about 1:1, around 37 trillion each! That your body consists of more microbial cells is a myth.
Give five examples of sources for ancient microbiomes (paleomicrobiological analysis).
There are many sources for ancient microbiomes:
- Soil, sediment, environmental samples
- Animal or human bone remains
- Ice cores
- Stone tools
- Pottery
- Rock paintings
- Dental calculus
- Coprolites
Although a lot of modern contamination, microbes are everywhere!
There are old and new methods for detecting ancient microbes, name and describe the five old methods for microbial aDNA and name advantages and challenges.
- Historical accounts and records: Can help with determining etymologies and provide historical context but variation in translation and interpretation and limited accuracy in description and causative agents. Hard to determine what disease it was.
- Culturing: Good for understanding metabolic capabilities of ancient microbes. Only works if the microbes are still “alive”, hard to “resuscitate” and not all microbes can be cultured, big contamination risk - don’t want to spread again.
- Microscopy: Used by Robert Koch finding the tuberculosis bacillus or Louis Pasteur finding the plague bacillus. Good because there is no need to isolate or culture the microbe beforehand. But, difficult to distinguish between different microbial species.
- Immunohistochemistry: Using antibodies for detection, need to know what you are looking for, for example used to identify rabies in a persons blood.
- Genetic approaches using PCR: Specific detection and can identify a variety in one sample. Needs primers so it only targets specific DNA sequences of a species - need to know what you are looking for and biased. Also contamination risk.
In summary, none of the old methods are adapted to full-scale ancient microbiome research
What four new methods are up and coming in microbial aDNA research? Advantages/challenges?
Metagenomics, metaproteomics, metabolomics and lately also metatranscriptomics. Massive amounts of data for all. Highly dependent on reference databases, which are not very good yet for most, but for metagenomics they are decent.
- Metagenomics: allows for specific detection of microbes and strains, also phylogeny. Enables comparison between ancient microbiome types but misses viruses with an RNA genome. Current gold standard in microbiome aDNA research.
- Metaproteomics proteomics and metabolomics are not capable of species determination (yet) but metaproteomics informs on much older samples because proteins seem to preserve better than DNA.
- Metatranscriptomics: Use of ancient RNA. For the moment in its infancy, RNA is more fragile than DNA and is more rapidly degraded. However most of the viruses carry an RNA genome like coronaviruses for example.
Take home message: Can be good to combine these methods depending on the question
What are the use cases of paleomicrobiological analysis (microbial aDNA)?
- Determining causative pathogen for historical disease outbreaks, e.g. yersenia pestis (black death)
- Understanding ancient human diet and environment
- Paleopathology: e.g. Life-long bacterial diseases affecting bones (tuberculosis, brucellosis, leprosy and syphilis)
- Finding emergence of pathogens and what happened, useful in future epidemics: e.g. we find more pathogens and outbreaks from the start of the Neolithic - people started to live in close proximity to each other and animals.
What are the main obstacles in ancient metagenomics?
- Contamination: aDNA in general is very low in abundance, compared to modern contamination, even worse for ancient microbes as most of the modern contamination is microbes - authentication important! Enrichment of authenticated aDNA helpful to detect ancient microbes.
- DNA damage: Fragmented, deamination, need good coverage.
- Database bias: The number of reads uniquely mapped to the correct organism decreases with larger reference databases. It can be a good idea to use smaller reference libraries.
- Misalignment to conserved regions across species: Also related to reference genome databases, some regions are conserved across all eukaryotes, so it can be a good idea to remove these from the alignment to get a stronger signal to a specific species.
How can we detect a microbe among the DNA sequenced? Provide three possible methods. Which is the best?
To detect identify a microbe from DNA sequence data, we can use three methods:
- Classifier (high sensitivity): Create a reference database and cut the genomes of these into short fragments ~31-35 bp long. Then do the same to the DNA sequence data unique K-mers: short DNA sequences ~31-35 bp long and then the reads are mapped to the reference genomes and assigned to a taxa. If many K-mers map to a reference and the coverage is high = good classification that is used for identification. Few K-mers and low coverage = bad classification.
- Aligner (High specificity): Uses an algorithm to align the sequenced reads to the reference genome of the species from which they came, this helps to bin different species from a pooled sample and helps to identify what is present.
- De novo Assembler: Recreate the genome of a microbe from scratch. This method could be really useful to discover unknown or now extinct ancient microbes. This method takes time and requires high amounts and good quality aDNA. It’s promising but limited to a certain number of samples.
Using a combination is a great way, e.g. aMeta (ancient Metagenomic profiling workflow): First use a classifier against a very large database, then filter the results and create a custom database based on it and align the samples to it. The alignment is a necessary step but using previous results to create a custom database make it much faster, more specific and computationally-efficient. Minimizes false positives.
What measures can be taken to filter the data and make the identification faster and more efficient?
- Exclude potential soil contamination by ignoring soil-dwelling microbes.
- Check ancient status
- Check potential misalignment from sister taxa
How can studies on coprolites be useful?
Coprolites was used to show the diet of iron age humans: Found microbes that are used in todays process of making beer and blue cheese, indicating that these foods were consumed already back then.
Coprolites can be useful for studying the ancient gut microbiome, especially eukaryotic parasites since they don’t enter the bloodstream and are unlikely to reach the teeth. However, they do not preserve DNA very well because they do not have a protective surface and are more contaminated.
How can studies on dental calculus be useful?
A study on dental calculus of humans from 6000 BCE and forwerds found the earliest detection of S. mutans during the bronze age 2000 BCE (Major contributor of dental plaque formation) and P. gingivalis in the beginning of Mesolithic 5500 BCE (Key pathogen in periodontitis - inflammation in gums, tooth loss) which could have been due to the shift to eating more carbs with farming.
Useful in diet determination between species.
What is Paleopathology?
The study of diseases and injuries in ancient individuals.
What sample types are there for pathogen detection?
- Coprolites for parasites or bacterial dysbiosis
- Dentin of teeth for viruses and pathogenic bacteria causing bacteremia (entering the bloodstream)
- Bone showing lesions (for chronic diseases)
- Calcified nodules
- Mummies: e.g Bishop Winstrup in Lund 1679, did indeed die from tuberculosis.
Note: petrous bones are not the best when it comes to pathogens! pathogens have less change of entering dense bones.
What use cases lie in the future for ancient microbiomes?
The future:
- Investigate gene evolution of pathogens: ymt gene: gene that facilitates the transmission of Yersinia pestis via fleas, This gene only appears between the late Neolithic and the late Bronze Age. How was Y. pestis transmitted before?
- Follow the repartition of a pathogen through time and space: e.g. Yersinia pestis have caused several epidemics, evidence that it spread to murmeldjur which functioned as a reservoir and then came back in a second epidemic.
- Pangenomes: discover genes that are not in the reference genome of that species today
- Studying ancient variants that were more susceptible to a certain pathogen, retained today? fixed? good for precaution.
There are mainly six areas/questions that are answered at a forensic/legal genetics lab, which? Are they common/uncommon?
Common:
- Identity testing: Match traces found at the crime scene with suspects.
- Relationship testing: Establish the biological relationship between two (or more) individuals. Often court cases like paternal testing.
- Missing person identification/DVI (disaster victim identification): Identification of human remains (accident, disaster etc).
Uncommon:
- Details about a biological sample: Tissue? Age (time since deposit)? Time of deposit (day/night)? Basically things that can be helpful in a police investigation.
- Details about an unknown donor Eye, hair, skin color? Age? Ancestry?Relatives? Species?
- Molecular diagnostics in forensic medicine: Cause of death due to genetic variants? Genotype vs phenotype, Ability to metabolize drugs
The three common questions answered by forensic labs are answered in the same way, how?
Identity testing, relationship testing and missing person identification are all based on the uniqueness in the number of Short Tandem Repeats (STRs) at 15-25 sites. Through PCR amplification of these 15-25 sites, and then running them on a gel, one can establish STR profiles from the different individuals/ samples involved in the case (fragment length analysis). These profiles can then be compared to related individuals that will also share these markers, more shared markers the more related they are to the person whose profile it is.
The results are probability based, and the analysts report the findings and then the authorities/court need to make the decision on how to use it.
Why do forensic analysis (mostly) use decades old DNA techniques?
Older methods have been around for a long time and have been repeatedly tested for accuracy:
- Highly acceptable in courts
- Based on ton’s of scientific and validation studies
- Forensic DNA databases (offender-, crime scene sample-) contain millions of STR profiles
- Consensus STR marker set, used all over the world - useful for international purposes.
The more uncommon questions answered by forensic genetics requires a bit of a different approach, what differs?
The common or easy questions usually only require very little genetic info to be answered, e.g. Legal question Is AF the biological father of the child? H1: AF is the father of the child, H2: AF is unrelated to the child. To answer this, 15-21 STRs is enough.
If the question is more vague with multiple outcomes or is exploratory, more genetic data is needed, for example: Is there any genetic relationship between Man 1 and Man 2? This requires between 1,000 –600,000 SNPs to be answered. So a big difference in complexity and resources.
Rättsmedicinalverket handles genetic questions “inside the body”, what four areas of expertise do they have?
Four areas of expertise:
–Forensic psychology
–Forensic medicine
–Forensic toxicology
–Forensic genetics
What biological material and how much is needed to obtain a DNA profile of good quality?
Amount of sample to obtain a DNA profile of good quality:
–DNA from ~100-1000 cells
–Someμl blood
–Some mm2 filter paper with saliva
–Somemm3 soft tissue
–0.5 g bone powder
–Some straws from toothbrush
–Some ml urine
Forensic labs can have an “Elimination database”, what is this used for?
An elimination database contains genetic data from employees, service technicians, cleaning personnel who may come into contact with samples and the sample environment. When a sample has been produced, this can be checked against this database to rule out contamination from them.
Forensic labs sometimes do species determination, why and how?
Species determination can be useful in cases of illegal hunting or to see whether remains are human (if that is not clear). For vertebrates this is usually done through amplification of mtDNA markers (e.g 16S rRNA mtDNA) and then sanger/pyro sequencing. Cheap, quick, easy.
