Use MAFFT (http://mafft.cbrc.jp/alignment/server/), it is slower than the programs you mention, but better! the output will need less manual aligning (fasta input, standard parameters are ok). For DNA I would use the qinsi model (slowest) as this considers secondary structure, but you can also review the other algorithms (the default may also gvie you an ok alignment). Then after a quick manual align (MEGA, MACCLADE, Mesquite) I would run it through Gblocks (http://www.phylogeny.fr/version2_cgi/one_task.cgi?task_type=gblocks) under the least stringent parameters (small final block, gap positions in final block and less strict flanking), to exclude poorly aligned positions and divergent regions from subsequent phylogenetic inferences(fasta input as I recall). OK, now you have an finished phylogeny friendly alignment! Then use modeltest to determine the best evolutionary model, but it should be GTR, then run using RAxML for ML or Bayes/phylobayes for Bayesian. You can run the phylogeny programs at the University of Oslo Bioportal (http://www.bioportal.uio.no/) and it also has tutorials to help. Hope this helps and good luck.

MEGA 5.1 was very useful for me. Additionally, if ITS 2 is your gene of interest, the ITS 2 database (http://its2.bioapps.biozentrum.uni-wuerzburg.de/) is an excellent resource.

Hi Chris

See the attached paper for a comparison among different tools for MSA, their pros and cons...it may be useful to you..

bye

I've recently published step-by-step protocol on alignment and phylogenetic analysis using MEGA and Geneious at nature protocol here:

http://www.nature.com/protocolexchange/protocols/2740

Hi Chris

what sequences are you using? are they protein coding or non coding ..this can make a huge difference in the analysis..i do not think there is any way of doing it simpler..I would also be interested in knowing the shortcut...but i sincerely doubt there is any...cheers

I will recommend BEAST

MEGA software. You can download from the internet freely.

Patrice Corneli mentioned the difficulty of getting the sequence files into a standard format. One useful tool I have found is the alignment transformation environment (ALTER) http://sing.ei.uvigo.es/ALTER/ which allows you to covert between formats. Neat tool.

Since you are at Florida State, you should have access to their cluster to streamline this process. All you need to do is have all of your files in fasta format, and then write a simple perl or python script to write shell scripts and submit each job individually. Given you have access to the processors, you should be able to complete your hundreds of alignments in the same amount of computation time for a single job. I recommend using MUSCLE, the only issue is that you'll need another script to reformat the output fasta files that removes the weird line breaks that MUSCLE likes to make.

Forget any manual alignment (unless you have compelling proof-of-concept for new approaches need) or profile-profile methods and accept that there will be some error. If alignment is really messy and you have coding sequences, you can translate the sequences to amino acids and map these sequences back to codons. There is a perl script called Translator X that can do this for you, or you'll have to develop the code yourself. I've had trouble with Translator X in the past when I let it guess the reading frame, but it usually gets it right- just peek in a few of the files to get an idea of its performance.

I recommend contacting your support staff for the cluster and they should be able to help you get all of this off the ground. As far as changing formats for other utilities, you'll want to script these; otherwise, things will take a while to do them one by one. There is usually somebody around who doesn't mind helping others with these problems, especially when it allows for collaboration, I recommend you find them in your department if you're not too comfortable with some of the programming. As Patrice mentioned, Haddock and Dunn's book is a great place to start for most of these ideas too.

Hi Chris,

There is no one solution. It depends on the organism and the gene(s) you are using. The following paper of mine gives details of the procedures I used to produce a simple gene tree based on 2 mitochondrial genes using time calibration with BEAST. Also mentioned are the alignment and display software. I hope you find this useful.

http://www.sciencedirect.com/science/article/pii/S0022098113000695

Good luck,

Ian G.

For how to use MEGA, there was a great article by Barry Hall in MBE recently: http://mbe.oxfordjournals.org/content/30/5/1229.long

If you're going to infer phylogenies using many loci I also recommend reading Arndt von Haeseler's "Do we still need supertrees?" (http://www.biomedcentral.com/1741-7007/10/13). It's a good commentary on how to approach this kind of work.

I don't agree with Patrice. Especially if you have a large number of sequences, manual editing is a nightmare. Check out MACSE for alignment, at the following link.

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0022594

Good luck

Ian G.

Use MAFFT (http://mafft.cbrc.jp/alignment/server/), it is slower than the programs you mention, but better! the output will need less manual aligning (fasta input, standard parameters are ok). For DNA I would use the qinsi model (slowest) as this considers secondary structure, but you can also review the other algorithms (the default may also gvie you an ok alignment). Then after a quick manual align (MEGA, MACCLADE, Mesquite) I would run it through Gblocks (http://www.phylogeny.fr/version2_cgi/one_task.cgi?task_type=gblocks) under the least stringent parameters (small final block, gap positions in final block and less strict flanking), to exclude poorly aligned positions and divergent regions from subsequent phylogenetic inferences(fasta input as I recall). OK, now you have an finished phylogeny friendly alignment! Then use modeltest to determine the best evolutionary model, but it should be GTR, then run using RAxML for ML or Bayes/phylobayes for Bayesian. You can run the phylogeny programs at the University of Oslo Bioportal (http://www.bioportal.uio.no/) and it also has tutorials to help. Hope this helps and good luck.

I agree with Russell Orr. One thing to add... when I was using MAFFT online, the number of sequences was limited at 500 if I remember correctly. The solution was to use MAFFT locally on your computer. That worked very nicely and MAFFT is one program that molecular evolution workshop prefers (see http://evomics.org/learning/bioinformatics/multiple-sequence-alignment/). I see that you are at around 100 sequences, but it might be somewhat faster if you run it locally. Enjoy.

I have used gblock too, it is pretty cool. Thanks everyone.

I have been using MAFFT on a linux machine. Seems fast and accurate for many sequences. Will be learning about SATé, PASTA, BAli-Phy, and UPP at the Evolution 2014 conference.

We published something of a walkthrough on this:

https://www.researchgate.net/publication/235601668_Incorporating_molecular_data_in_fungal_systematics_a_guide_for_aspiringresearchers?ev=prf_pub

Article Incorporating molecular data in fungal systematics: A guide ...

Mesquite now has the means to run mafft, opal, Clustal, and muscle. Pretty neat.

 Hellow Mr. Bast,

Can you please mail me your above mentioned nature protocol paper (http://www.nature.com/protocolexchange/protocols/2740) on [email protected]?  

You can try this. 

It is an amazing article...

Thank you Felix

Solfware MEGA

MAFFT is my preferred option now.

Morning. You can check this very good paper. Let me know if it works please. 

https://www.nature.com/protocolexchange/protocols/2740 

Good luck!

If you are using specific primer then you need to match both strands by making reverse compliment then by using cap contid assembly in bioedit software you will get final seq for analysis. The edited sequences should be added to MEGA software for further analysis firstly u need to run clustal W for alignment then u can do any analysis based on your study