#bioinformatic

let’s talk a little bit about ancient DNA, shall we? :) 

As a logical epidemiological consequence after European contact, several new pathogens were brought to America. To this followed many epidemic outbreaks which had devastating consequences for local, native populations. Studying these diseases and its causal agents pose a great challenge for scientists. On one hand, while some outbreaks were recorded, propose a specific pathogen is very difficult to do when only records of symptoms are available. Adding to this, clinical presentations of certain diseases may have evolved or changed over the years. As a consequence most pathogenic agents that caused these outbreaks still unidentified to date. Thus researchers need to rely on other resources if they are to disentangle these puzzling outbreaks. Here’s where things get interesting because once again DNA tools make their heroic appearance.

Of course the solution didn't come overnight, and first and foremost one needs to have a source of DNA right?. 

One of the most interesting cases is the so-called cocoliztli colonial epidemic in Mexico which took place between 1545-1550. This epidemic caused the death of approximately 80% of the native population. Just yesterday Nature Ecology and Evolution published the paper of the international team, led by researchers from the Max Planck Institute for the Science of Human History (MPI-SHH), Harvard University and the Mexican National Institute of Anthropology and History (INAH). They made the most of the only known cemetery linked to this particular outbreak located at Teposcolula-Yucundaa (Oaxaca, México), in order to investigate the pathogen involved in this outbreak. 

Samples were obtained from teeth, specifically from the crown pulp chamber. After processing according to DNA archaeological bone protocol and shotgun sequencing the samples, they applied and actually pioneered in a new bioinformatic-metagenomic analysis tool called MALT (Megan ALignment Tool). 

The advantage of malt-build (version 0.3.8) is the inclusion of all complete bacterial genomes in the reference database, which in turn allows identifying both pathogenic and non-pathogenic bacteria. 

“A key result of this study is that we were successful in recovering information about a microbial infection that was circulating in this population, and we did not need to specify a particular target in advance” explains Alexander Herbig, also of the MPI-SHH and co-first author of the study. In the past, scientists usually targeted a particular pathogen or a small set of pathogens, for which they had a prior indication. “This new approach allows us to search broadly at the genome level for whatever may be present,” added Johannes Krause, director of the Department of Archaeogenetics at the MPI-SHH and last author of the study. Kirsten Bos, also of the MPI-SHH, adds, “This is a critical advancement in the methods available to us as researchers of ancient diseases – we can now look for the molecular traces of many infectious agents in the archaeological record, which is especially relevant to typical cases where the cause of an illness is not known a priori.”

They went further performing various interesting and very comprehensive methodologies. For example about the phylogeny of Salmonella enterica using 5 ancient genomes. As well as screening for presence/absence of virulence factors using previously identified effector genes. The paper presents genome-wide data from 10 individuals for Salmonella enterica subsp. enterica serovar Paratyphi C, the bacterial cause of the enteric fever. Which they propose as a strong candidate aetiology for the epidemic cocoliztli outbreak at Teposcolula-Yucundaa.

Read the whole paper here

Quotes from authors were obtained from this note from the Max Planck Institute for the Science of Human History here.

Read some background (in Spanish). 

Continuing the theme of biology computer games, Phylo is an online game where you try to find the optimal alignment between multiple sequences.

Sequence alignment is an integral part in the modern determination of phylogeny, whether the sequence is from DNA, RNA, or proteins.  For the most part, modern sequence alignment is done through complex algorithms like ClustalW, MAFFT, and MUSCLE.  These use complex matrices and scoring systems with point values for matches, mismatches, gaps, and gap extensions.    Having done several pairwise alignments by hand, they are a huge pain in the ass, even for as few as 10 bases. Since the average human gene is 3 kilobases long, it's exceedingly impractical to align sequences by hand, especially when you are aligning more than two sequences.  

Phylo starts you with two sequences and begins to add more as you progress.  And while aligning sequencing by matrices is enough to drive a person mad, it's actually pretty entertaining to do it by sight and simple logic. 

All of the DNA sequences you are given to align are actual sequences from human genes that may be related to genetic diseases available in the UCSC Genome Browser.

A paper about Phylo was published PLOS ONE in January 2012.  PLOS ONE is one of those lovely scientific journals that doesn't require a subscription to read articles so here, happy reading!