You may remember that a while ago I blogged about a research group that found a way to produce biogas reactor using the content of a stomachs cow. One of the questions, I had back then was what kinds of bacteria were in the cow's stomach fluids?
Today, we'll look at a massive Science article from January 28 with the title "Metagenomic Discovery of Biomass-Degrading Genes and Genomes from Cow Rumen" from the Rubin Lab with Matthias Hess and Alexander Sczyrba as the primary authors recently hit the science world, and gives us that answer.
What did they do?
The research group figured out a way to create a sealable opening into a cows stomach. They used this opening to insert bags containing switch grass (a target for the production of second generation biofuels) into the stomach of the cow which they were also able to remove after a given amount of time. They then broadly asked what kinds of bacteria are living in the cow rumen by sequencing the DNA extracted from the bags which they incubated in the cow's stomach. Because DNA pieces obtained through the current massively parallel sequencing technologies produce rather short fragments, it is initially very hard to assign a given piece of DNA to a given bacterial genome. Thus, the DNA soup obtained from this massive sequencing effort is called a metagenome. Although accurate assignment may not be possible with metagenomic analysis, there are still many questions one can ask.
The motivation behind the sequencing effort was to identify new possible enzymes that are capable of breaking down different forms of carbohydrates (cellulose in particular). They group was able to find many, many candidates. They tested a subset of these enzymes (90 candidates), and found that 53 of those enzyme candidates indeed had some sort of activity in their panel of 9 different carbohydrates - a very significant enrichment for proteins with carbohydrate activity. Although other papers have identified a slew of these enzymes already, it should be noted that their methologies allowed for identification of a whole set of new candidates by not relying on sequence identity but more so on the presence of certain typical functional domains.
Lastly, it should be again noted that metagenomics is messy because it is hard to tell which of the billions of short DNA sequencing reads belong to which species. Yet most remarkably, the research group was able to assemble and propose 446 draft genomes corresponding to 446 proposed new bacterial species. The group was able to do so because the coverage (how many times a particular region of the genome is captured by a snippet of DNA) was very high (53x). High coverage leads to a high likely hood that fragments overlap. These overlaps were exploited to assemble "scaffold assemblies" (essentially a larger continuous piece of DNA). Using something called TNF (tetranucleotide frequencies) and read coverage as a measure of abundance, the group was able to then bin individual contigs which is how they came up with the number 446. This number likely is an overall underestimation of the actual number of different bacterial species out there because there is a bias against rare species.
Why is this article important?
Firstly, there is a lot of technical detail hidden in this article that suggest a way to approach metagenomic analysis of such scale. Secondly, this article is a treasure box full of new enzymes that can be used to make biofuels production more efficient and thus more affordable. I suspect that any biofuels company out there will probably study this article in great detail. And lastly, the ability to culture a bacterial species had previously been a condition to study it including trying to sequence its genome. In this case, 446 brand new draft genome sequences were inferred from bacteria that had never been cultured before!
On a personal observational note, I find the recent trend in the articles that appeared in Nature and Science including our previous journal club in which we looked at the ability of two bacterial species to pass off electrons from one to another interesting because it reflects a transition in the field in which the importance of biological research is expanding out to other areas not limited to biomedical research. Exciting times are ahead of us!
Today, we'll look at a massive Science article from January 28 with the title "Metagenomic Discovery of Biomass-Degrading Genes and Genomes from Cow Rumen" from the Rubin Lab with Matthias Hess and Alexander Sczyrba as the primary authors recently hit the science world, and gives us that answer.
What did they do?
 |
| Image of Cow with Fistula. Source: Scientific American |
The motivation behind the sequencing effort was to identify new possible enzymes that are capable of breaking down different forms of carbohydrates (cellulose in particular). They group was able to find many, many candidates. They tested a subset of these enzymes (90 candidates), and found that 53 of those enzyme candidates indeed had some sort of activity in their panel of 9 different carbohydrates - a very significant enrichment for proteins with carbohydrate activity. Although other papers have identified a slew of these enzymes already, it should be noted that their methologies allowed for identification of a whole set of new candidates by not relying on sequence identity but more so on the presence of certain typical functional domains.
Lastly, it should be again noted that metagenomics is messy because it is hard to tell which of the billions of short DNA sequencing reads belong to which species. Yet most remarkably, the research group was able to assemble and propose 446 draft genomes corresponding to 446 proposed new bacterial species. The group was able to do so because the coverage (how many times a particular region of the genome is captured by a snippet of DNA) was very high (53x). High coverage leads to a high likely hood that fragments overlap. These overlaps were exploited to assemble "scaffold assemblies" (essentially a larger continuous piece of DNA). Using something called TNF (tetranucleotide frequencies) and read coverage as a measure of abundance, the group was able to then bin individual contigs which is how they came up with the number 446. This number likely is an overall underestimation of the actual number of different bacterial species out there because there is a bias against rare species.
Why is this article important?
Firstly, there is a lot of technical detail hidden in this article that suggest a way to approach metagenomic analysis of such scale. Secondly, this article is a treasure box full of new enzymes that can be used to make biofuels production more efficient and thus more affordable. I suspect that any biofuels company out there will probably study this article in great detail. And lastly, the ability to culture a bacterial species had previously been a condition to study it including trying to sequence its genome. In this case, 446 brand new draft genome sequences were inferred from bacteria that had never been cultured before!
On a personal observational note, I find the recent trend in the articles that appeared in Nature and Science including our previous journal club in which we looked at the ability of two bacterial species to pass off electrons from one to another interesting because it reflects a transition in the field in which the importance of biological research is expanding out to other areas not limited to biomedical research. Exciting times are ahead of us!
Comments
Post a Comment