User:Westrick.36/sandbox

This is Dayna's sandbox.

Wikipedia Assignment

https://en.wikipedia.org/wiki/Talk:Long_branch_attraction

Sentence to the article:

Until recently, long branch attraction was considered hypothetical due to insufficient evidence. However, today many different factors are taken into account in confirmation of the distance between two species (Bergsten 2005).


 * Bergsten J. (2005): "A review of long-branch attraction". Blackwell Publishing [cited 2014 Oct 1] 21(2):163-193. Available from: http://onlinelibrary.wiley.com/doi/10.1111/j.1096-0031.2005.00059.x/pdf

Suggestions:

I think something important to add to the article would be the methods of identifying species that exhibit long branch attraction. The idea is recent so not many people will know how to tell apart the species that exhibit similar characteristics but have a very ancient ancestor.

Examples of species that exhibit long branch attraction would also be welcomed because examples would help to back up the idea.

It might also be helpful to insert a morphology tree to be able to show the relationship between species that exhibit long branch attraction.

My Addition to the Wikipedia Page
https://en.wikipedia.org/wiki/Long_branch_attraction

Example

One example of this phenomenon is the relationship between four skippers (butterflies): Agathymus mariae, Ancyloxpha numitor, Thorybes pylades, and Pyrrhopyge zenodorus. When comparing these species scientists used a multitude of procedures to compare to one another in order to get the most accurate results.

They began with analyzing a certain length of DNA in each species. Compared side by side they counted the matching nucleotides in each strand and came up with a phylogenetic tree based on the similarity shared between each DNA strand. This resulted in a tree supporting the close relationship between A. mariae and P. zenodorus.

The nest step in the process involved another reconstruction method, distance-based. The amount of expected changes within each given DNA sequence was estimated. The species with similar amounts of changes were grouped together and were calculated to have a bootstrap value of 80%, also supporting tree 1.

The next method used in this procedure is called Maximum likelihood. So far in the data analysis, the trees have been in parsimony, meaning they have been the simplest forms. However, maximum likelihood is a process that takes into account what changes are the most likely to occur. It is not necessarily the easiest tree but is one that is the mostly likely to statistically occur. The maximum likelihood tree supports at tree linking A. numitor and A. mariae. This is the first method to have results that conflicts with that of the previously executed methods.

Another method to compare results is called Bayesian method. This method is very similar to maximum likelihood. It deals with the statistical data and creates a tree that represents the most likely occurrence. It differs from maximum likelihood in that it predicts how likely it would happen in the future. In this data set analysis, the Bayesian method resulted in a tree that also supports the close relationship of A. numitor and A. mariae.

When all this data is gathered and compared, we find that the second relationship is the most logical relationship. This experiment with skippers supports the importance of deciphering all the data before concluding that a certain tree is the correct one. Morphological traits are very important aspects of constructing trees, but parsimony is not always correct. It is helpful in using a few methods to determine an accurate tree (Grishin 2009).

Reference


 * Grishin, Nick V. "Long Branch Attraction." Long Branch Attraction. Butterflies of America, 17 Aug. 2009. Web. 15 Sept. 2014. .

FINAL DRAFT STARTS HERE
The placing of species onto a phylogenetic tree is an elaborate process. With the basics, such as morphological traits, one is able to attempt to deduce the relationships between a handful of species. However, morphological traits are almost never enough to fully grasp the true relationships between species. Constructing a phylogenetic tree is a process that involves much data gathering and interpreting. The best way to understand these relationships is to analyze the core of a species, such as examining nuclear, mitochondrial, and chloroplast protein coding genes. Furthermore, examining these aspects cannot always guarantee a perfect phylogenetic tree. When creating these trees, the use of parsimony (finding the simplest tree) and maximum likelihood are some of the most common ways of inferring the best branching relationships. However, a concept known as long branching attraction can come into conflict with these methods. Long branching attraction is the independent evolution of two non-sister branches that acquire similar characteristics (analogous traits), causing the two species to be placed close together on the phylogenetic tree. This idea is a fairly new theory that is still evolving and is constantly under scrutiny. With the aid of data gathering and other methods, long branching attraction can be labeled as an important part of determining phylogenetic trees, and within this paper I will discuss some examples of this phenomenon. The concept of long branching attraction is fairly new and conflicts with previously identified aspects of a phylogenetic tree. Scientists have identified 112 cases in which long branching attraction has been a problem in the sister taxa of previously constructed trees. Therefore, they have attempted to reconstruct new, more accurate trees based on methods I will discuss later on in the paper. One probable sign indicating the possibility of long branching attraction is the discrepancy between morphological traits and molecular data. If two trees were constructed, between data sets of morphological traits and molecular traits, one would be able to compare the resulting trees and determine if there is a problem with the categorization of certain species. Clashes, such as these, arise when two different species evolve individually but similarly. Species that acquire these analogous traits can make it extremely difficult to be categorized into correct clades. Long branching attraction is a phenomenon of molecular data. For example, in a nuclear sequence, if an “A” or “ala” is obtained at a certain point in two separate species, they will look identical. This random similarity between two non-sister branches is what causes the long branching attraction occurrence (Bergsten 2005). Using parsimony to group these two species would then put them on a tree as sister branches. This grouping would be incorrect because these species did not diverge from a common ancestor, they only evolved, by chance, in a similar manner. There can be many different explanations as to why two different species would evolve in a similar manner, one being the evolution of a trait in order to solve an environmental problem. An example of this would be the length of tails on certain birds. A longer tail is used by birds for better aerodynamics, but when two different species of birds obtain this longer tail, it does not necessarily mean that they are sister taxa. Both species of birds found the long tail beneficial in their environment. This analogous trait is a reason long branching attraction occurs.

One example of this phenomenon is the relationship between four skippers: Agathymus mariae, Ancyloxpha numitor, THorybes pylades, and Pyrrhopyge zenodorus. When comparing these species scientists used a multitude of procedures to compare to one another in order to get the most accurate results. They began with analyzing a certain length of DNA in each species. Compared side by side they counted the matching nucleotides in each strand and came up with a phylogenetic tree based on the similarity shared between each DNA strand. This resulted in a tree supporting the close relationship between A. mariae and P. zenodorus (tree 1). The nest step in the process involved another reconstruction method, distance-based. The amount of expected changes within each given DNA sequence was estimated. The species with similar amounts of changes were grouped together and were calculated to have a bootstrap value of 80%, also supporting tree 1. The next method used in this procedure is called Maximum likelihood. So far in the data analysis, the trees have been in parsimony, meaning they have been the simplest forms. However, maximum likelihood is a process that takes into account what changes are the most likely to occur. It is not necessarily the easiest tree but is one that is the mostly likely to statistically occur. The maximum likelihood tree supports at tree linking A. numitor and A. mariae (tree 2). This is the first method to have results that conflicts with that of the previously executed methods. Another method to compare results is called Bayesian method. This method is very similar to maximum likelihood. It deals with the statistical data and creates a tree that represents the most likely occurrence. It differs from maximum likelihood in that it predicts how likely it would happen in the future. In this data set analysis, the Bayesian method resulted in a tree that supports the close relationships in tree 2. When all this data is gathered and compared, we find that tree 2 is the most logical relationship. This experiment with skippers supports the importance of deciphering all the data before concluding that a certain tree is the correct one. Morphological traits are very important aspects of constructing trees, but parsimony is not always correct. It is helpful in using a few methods to determine an accurate tree (Grishin 2009).

Another example of long branching attraction deals with Iguanas. The trees constructed from a morphological data set and a molecular data set were incongruent with one another. In the process of the reconstruction of the Iguanidae tree, long branching attraction was taken into account. The former trees had many differences, which posed the question, whether the morphological or molecular data set, or both, was incomplete and needed further study to support the separate phylogenetic trees. In this case all the information obtained in previous experiments were reanalyzed to determine where the conflict lies. This species analysis was conducted in a way similar to that of the experiment with skippers, as discussed before. One of the main differences when working with these species is that the amount of data used and the number of phylogenetic trees created was numerous. The results of using a larger amount of data is that the final tree will be more accurate than that of an experiment that uses only a few. The initial phylogeny of the Igunanas were found wrong due to the long branching attraction witnessed between the morphological and molecular traits. Using maximum likelihood the correct tree was derived from all the data collected (Wiens 2000).

Long branching attraction is found in some of the early eukaryotic evolution. The main method of grouping for early eukaryotes is to analyze the rDNA of species. However, recently there has been debate as to whether this data alone is an accurate way to classify these species. Addressing this issue with the idea that long branching attraction may be causing this inaccuracy is helpful. Knowing that a tree is incorrect based on the fact that the data currently supporting this tree may be too little will cause one to go back and take into consideration other morphological or molecular traits (Stiller 1999).

Long branching attraction is a new and rising idea in the world of biology. The idea that a sister branch of species may not be as related as we thought has brought to our attention the amount of data that actually is needed to create phylogenetic trees. Long branching attraction may not be totally proven, but like any hypothesis, it has been found to be the least refuted. When addressing the issue of long branching attraction, it is important to use as much data as possible. A tree is not always correct when just looking at the morphological traits, so we compare and contrast these trees with molecular data as well. The methods behind the data analysis is also important. Parsimony may be the easiest way to build a tree, but it may not always be accurate. The correct way to analyzing data is by using maximum likelihood and parsimony together. Combining all this data and the data analysis methods will increase the likelihood of finally creating a tree that is the most accurate.

Work Cited
 * Bergsten, Johannes. "CladisticsVolume 21, Issue 2, Article First Published Online: 31 MAY 2005." A Review of Long-branch Attraction. Blackwell Publishing, 14 Feb. 2005. Web. 16 Sept. 2014. .
 * Stiller, J. W., and B. D. Hall. "Long-Branch Attraction and the RDNA Model of Early Eukaryotic Evolution." Molecular Biology and Evolution 16.9 (1999): 1270-279. Oxford Journals. Web. 15 Sept. 2014. .
 * Grishin, Nick V. "Long Branch Attraction." Long Branch Attraction. Butterflies of America, 17 Aug. 2009. Web. 15 Sept. 2014. .
 * Stefanović, Saša, Danny W. Rice, and Jeffery D. Palmer. "Long Branch Attraction, Taxon Sampling, and the Earliest Angiosperms: Amborella or Monocots?" BMC Evolutionary Biology. N.p., 28 Sept. 2004. Web. 15 Sept. 2014. .
 * Kück, Patrick, Christoph Mayer, Johann-Wolfgang Wägele, and Bernhard Misof. "Long Branch Effects Distort Maximum Likelihood Phylogenies in Simulations Despite Selection of the Correct Model." PLOS ONE:. N.p., 09 May 2012. Web. 15 Sept. 2014. .
 * Wiens J, Hollingsworth B. 2000. War of the Iguanas: Conflicting Molecular and MorphologicalPhylogenies and Long-Branch Attraction in Iguanid Lizards. Oxford Journals [Internet]. [cited 2014 Oct 24] 49(1):143–159. Available from: http://sysbio.oxfordjournals.org/content/49/1/143.full.pdf+html
 * Telford M, Copley R. 2005. Animal Phylogeny: Fatal Attraction. Science Direct [Internet]. [2005 Apr 26, cited 2014 Oct 24] 15(8). Available from: http://www.sciencedirect.com/science/article/pii/S0960982205003714
 * Lartillot N, Brinkmann H, Philippe H. 2007. Suppression of long-branch attraction artefacts in the animal phylogeny using a site-heterogeneous model. BioMed Central [Internet]. [2007 Feb 8, cited 2014 Oct 24] 7(Suppl 1):S4. Available from: http://www.biomedcentral.com/content/pdf/1471-2148-7-S1-S4.pdf