Phylogenetic trees traditionally involved genes. We are able to see phylogenetic trees resolving evolutionary relationships between gene and protein level in eukaryotes. 16S RNA sequences are used for classification and phylogeny among bacteria. Phylogenetic trees are a old concept, but they come with a new twist as now whole genome data is available. Can we resolve phylogenetic relationships between genomes of microbes. Is there a metric for such a study rather if this is an open question still ?!
Dear Sathya,
We have been able to investigate the phylogenetic relationships between bacteria, at a number of different taxonomic levels, using 16S RNA sequences for sometime, and it is really no different from similar analyses of eukaryote lineages using 18S RNA sequence data. However, there are some questions that need to be considered.
First, as more isolates are sequenced, the ability to differentiate between organisms based solely on the 16S RNA sequences becomes increasingly difficult. For example, the pseudomonads are long recognised as containing significant diversity at the species level, and many strains with clear phenotype differences cannot be differentiated using 16S RNA sequences. The way forward is to start to use ocncantenated sequences including a number of house keeping genes that are conserved amongst the group of organisms being investigated. For the pseudomonads, seven of these seem to allow better differentiation of isolates within species and species complexes. However, as whole genome sequences are becoming more readily available, the number of conserved genes that could be used is increasing, and whole genome comparisons are becoming more common.
Second, it is increasingly clear that phylogenetic tress based solely on 16 / 18S RNA sequences do not answer all the questions we would like to examine. It is clear that for many phenotypes of interest genes and operons might be transferred across taxonomic branches by horizontal (lateral) gene transfer (HGT). This means that trees based on 16S RNA sequences start to lose meaning, as the phenotype of interest may not be strongly linked to the core 16S 'definition' of an organism. As a result, alternative tress can be constructed using genes of interest (or even ocncantenated genes as before) to produce trees that can be compared to 16S trees. Deviations in branch structure can then suggest HGT, which can then be examined by other techniques.
It is increasingly becoming obvious that bacterial genomes are composites, with small and large fragments being transferred within communities and over evolutionary time to result in hybrid genomes that may preserver or be rapidly lost. This itself is a very interesting area in bacterial evolution to consider, but at the same time, we still need a simple method to categories bacteria, and for this, 16S or concatenated house keeping gene tress are probably the best means of doing so.
Regards, Andrew
Dear Sathya,
We have been able to investigate the phylogenetic relationships between bacteria, at a number of different taxonomic levels, using 16S RNA sequences for sometime, and it is really no different from similar analyses of eukaryote lineages using 18S RNA sequence data. However, there are some questions that need to be considered.
First, as more isolates are sequenced, the ability to differentiate between organisms based solely on the 16S RNA sequences becomes increasingly difficult. For example, the pseudomonads are long recognised as containing significant diversity at the species level, and many strains with clear phenotype differences cannot be differentiated using 16S RNA sequences. The way forward is to start to use ocncantenated sequences including a number of house keeping genes that are conserved amongst the group of organisms being investigated. For the pseudomonads, seven of these seem to allow better differentiation of isolates within species and species complexes. However, as whole genome sequences are becoming more readily available, the number of conserved genes that could be used is increasing, and whole genome comparisons are becoming more common.
Second, it is increasingly clear that phylogenetic tress based solely on 16 / 18S RNA sequences do not answer all the questions we would like to examine. It is clear that for many phenotypes of interest genes and operons might be transferred across taxonomic branches by horizontal (lateral) gene transfer (HGT). This means that trees based on 16S RNA sequences start to lose meaning, as the phenotype of interest may not be strongly linked to the core 16S 'definition' of an organism. As a result, alternative tress can be constructed using genes of interest (or even ocncantenated genes as before) to produce trees that can be compared to 16S trees. Deviations in branch structure can then suggest HGT, which can then be examined by other techniques.
It is increasingly becoming obvious that bacterial genomes are composites, with small and large fragments being transferred within communities and over evolutionary time to result in hybrid genomes that may preserver or be rapidly lost. This itself is a very interesting area in bacterial evolution to consider, but at the same time, we still need a simple method to categories bacteria, and for this, 16S or concatenated house keeping gene tress are probably the best means of doing so.
Regards, Andrew
With the limitations commented by Andrew, 16S is still the simplest method to categorise bacteria. But there are genome based metrics, such as AAI and ANI. Genome sequencing is closing to be the ultimate microbial phylogenetic classification method. Here is one of the latest metrics: http://nar.oxfordjournals.org/lookup/doi/10.1093/nar/gkv657
I totally agree with the answers of my fellows here, Andrew and Kurth. I just would add this simple question/answer: can a single gene encode the whole genome information? of course not.. a typical 16s gene is about 1500bp, while corresponding genomes is at least one order of magnitude higher. Information encoded in phylogenetic gene markers only account for the vertical passing of information in biological organisms, but biology can pass genetic information horizontally too... especially in microorganisms. Hence, when you are going to tackle the evolution problem, more information is needed; hopefully with whole genome techniques, which will take into account both vertical and horizontal transfer of information.
Best wishes.
---sram
Thanks everyone for the great answers.
My focus is on genome metrics. We need to wait and see how they will be standardized.
IMHO you don't have to wait.. be active here and propose your own metrics
I would love to see proposals about that very interesting topic here
Best regards,
---sram
Follow my publication
you will get the answer.
https://scholar.google.co.in/citations?user=PSoYZ6MAAAAJ&hl=en&authuser=1
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology