Dear all,
Thanks a lot for your answers ! Here are the main references I obtained
from you on choice for phylogeny methods / models :
At first, some books :
1) Barry Hall. Phylogenetic Trees Made Easy: A How-To Manual Bellingham
Research Institute ; Third Edition (Bayesian methods should be
integrated in this version ; recommended twice) :
http://www.sinauer.com/hall/3e/
2) Page & Holmes. Molecular Evolution. A Phylogenetic Approach
(recommended twice)
http://www.amazon.com/Molecular-Evolution-Roderick-D-M-Page/dp/0865428891/ref=sr_1_5?ie=UTF8&qid=1326723824&sr=8-5
3) Hillis, Moritz, Mable. Molecular Systematics (recommended twice)
"If you read through these three in order (or at least 1+2), you should
get a solid foundation of understanding different methods and models
and how to choose among them. I believe #1 has been updated to include
Bayesian methods, but I'm not sure about #2+3."
[ Jaret Bilewitch ]
- The Phylogenetic Handbook A Practical Approach to Phylogenetic
Analysis and Hypothesis Testing, by P. Lemey, M. Salemi, A.M. Vandamme
(recommended twice)
http://www.amazon.com/Phylogenetic-Handbook-Practical-Approach-Hypothesis/dp/0521730716/ref=pd_sim_b_1
- Phylogenetics The theory and practice of phylogenetic systematics, by
E.O. Wiley,
- Inferring Phylogenies, Joseph Felsenstein :
http://www.sinauer.com/detail.php?id=1775
- (in french :) Concepts et méthodes en phylogénie moléculaire
Collection: Collection IRIS
Perrière, Guy, Brochier-Armanet, Céline
1st Edition., 2010, XII, 250 p.
http://www.springer.com/life+sciences/systems+biology+and+bioinformatics/book/978-2-287-99047-2
Then, some articles too :
- Kelchner SA, Thomas MA. 2007. Model use in phylogenetics: nine key
questions. Trends Ecol Evol. 22:87–94.
- Papers written by Nei and Saitou explaining advantages and
disadvantages of each type of method :
Saitou N, Imanishi T (1989) Relative efficiencies of the
Fitch-Margoliash, maximum-parsimony, maximum-likelihood,
minimum-evolution, and neighbor-joining methods of phylogenetic tree
construction in obtaining the correct tree. Mol. Biol. Evol. 6(5):514-525.
Takahashi K, Nei M (2000) Efficiencies of fast algorithms of
phylogenetic inference under the criteria of maximum parsimony, minimum
evolution, and maximum likelihood when a large number of sequences are
used. Mol. Biol. Evol. 17(8): 1251-1258.
Tateno Y, Takezaki N, Nei M (1994) Relative efficiencies of the
maximum-likelihood, neighbor-joining, and maximum-parsimony methods when
substitution rate varies with site. Mol. Biol. Evol. 11(2):261-277.
- And for bayesian inference (BI) :
Douady CJ, Delsuc F, Boucher Y, Doolittle WF, Douzery EJP (2003)
Comparison of bayesian and maximum likelihood bootstrap measures of
phylogenetic reliability Mol Biol Evol 20(2):248-254
Alfaro ME, Zoller S, Lutzoni F (2003) Bayes or bootstrap? A simulation
study comparing the performance of bayesian Markov chain Monte Carlo
sampling and bootstrappingin assessing phylogenetic confidence. Mol.
Biol. Evol. 20(2): 255-266.
Miscellaneous :
- Workshops on phylogeny (recommended twice) :
http://www.molecularevolution.org/
- In the documentation of the web server Phylemon2, the authors explain
a roadmap that is normally used for phylogenetic studies and seems very
interesting as a "lab protocol":
http://docs.bioinfo.cipf.es/phylemonwiki/doku.php?id=phylemon:roadmap
- Two programs on model choice within algorithm :
. for models of maximum likelihood (ML) or BI, there is a program called
jModelTest (http://darwin.uvigo.es/software/jmodeltest.html) that lets
you choose the best model that fits your data when performing the
phylogenetic tree in ML or BI.
. to select between alternative substitution models, there is
modelgenerator, available here (The paper is also listed there and they
discuss in the paper the whole idea of comparing models) :
http://bioinf.nuim.ie/modelgenerator/
Other advices :
- "Talks with people, since a practical experience is sometimes needed
in order to take a good decision. A secure approach remains a test of
different methods, to see their impact on obtained results"
[Céline Brochier ]
- " You will find , as I found myself when I was in your position
regarding phylogenetics methods, that no method or model it's
bulletproof, and the selection or the point of view that most of the
authors use to select the method are based on mere "personal
philosophies" or trends. If you read a paper of P. Goloboff by
example, it will talk wonders about Maximum Parsimony, but nowadays
everybody want to include Bayesian method, or if they don’t understand
quite the basics of the Bayesian theorem or application of the MCMC
algorithm, they will use Maximum Likelihood.
My recommendation is that if you are not quite sure of all the
assumptions on EVERY method or model, use the one that you understand
the most, or if you don't want to take that choice, analyze your
sequences, if they have a lot of Parsimonious sites it will be
advantageous to use Maximum Parsimony, or if you have a sequence very
long (and don’t have a lot of sequence aligned) you can use maximum
likelihood. Personally I don’t recommend using Bayesian method if you
are just starting in the application of Phylogenetics. "
[ Gabriel Golczer ]
Hope it helps ! (and it does, undoubtedly)
Thanks again to all of you (lot has not been cited here !)
All the best,
Gwennaël
-------- Message original --------
Dear all,
I am quite new in phylogeny and I know the very bases. I know some books
about phylogeny, but they all seem to present phylogeny methods and
distance matrices models without explaining how to make a good choice
between those methods and models (though it is interesting to compare
the results from different methods).
My question is thus: do you know some interesting references (in English
or in French) explaining practically how to select among phylogeny
methods and models depending on your data ?
Thank you very much in advance.
Regards,
Gwennaël
--
Gwennaël BATAILLE, PhD student - Teaching assistant
Earth and Life Institute
Université Catholique de Louvain
SST/ELI/ELIB
Bâtiment Carnoy
Croix du sud 4-5, bte L7.07.04
1348 Louvain-la-Neuve
BELGIUM
gwennael.bataille@uclouvain.be