The Muscle alignment

[NGS group]

Hi,

I am trying to use Muscle to do alignment of 12 bacterial genomes in Linux 64bit OS, which was installed by Virtual Box in Win 7. However, I got the problem showed as follow:

root@xxxx-VirtualBox:/home/xxxx/Downloads# muscle -in 12core_genome_nodash.fasta -out 12core_MUSCLE.afa -maxiters 2

MUSCLE v3.8.31 by Robert C. Edgar

http://www.drive5.com/muscle

This software is donated to the public domain.
Please cite: Edgar, R.C. Nucleic Acids Res 32(5), 1792-97.

12core_genome_nodash 12 seqs, max length 4249046, avg length 4248840
00:02:20 305 MB(-41%) Iter 1 100.00% K-mer dist pass 1
00:02:20 305 MB(-41%) Iter 1 100.00% K-mer dist pass 2
Killed11 7468 MB(-1004%) Iter 1 9.09% Align node

It seems that such problem would be due to the running out of memory. However, I assigned 16GB memory to my Virtualbox, which should be enough for MUSCLE as showed here (7468MB).

Could anyone tell me how to solve this problem?

Thanks a lot!

-Victor

[mcnelson.phd]

I'm wondering why you're trying to do whole genome alignments with muscle instead of something like Mauve or MUMmer which are designed specifically for such purposes?

[NGS group]

Because I need Phylip format file for building phylogenetic tree, which can not be produced by Mauve. I am not sure whether MUMer can generate the Phylip file. Do you know that?

[rhinoceros]

From MUSCLE's manual

Use –log or –loga and –*verbose and check the log file to see if there are any messages that give you a hint about the problem. Look at the peak memory requirements (reported in progress messages) to see if you may be exceeding the physical or virtual memory capacity of your computer.

[NGS group]

From MUSCLE's manual

Could you please show me the command line, since I am quite new on linux system? Thank you very much!

[rhinoceros]

Could you please show me the command line, since I am quite new on linux system? Thank you very much!

http://www.drive5.com/muscle/manual/index.html

http://www.drive5.com/muscle/manual/options.html

[SES]

My advice would be to align them with MUMmer, filter the alignment with the program delta-filter (if necessary), and then produce easier to parse formats with the show-coords and show-aligns programs. From there, it should take just a little scripting to get the format you want.

Alternatively, you may be able to create a pairwise alignment and add sequences one at a time with MAFFT, but I'm not sure how that will work for whole genomes. Either way, I'm afraid that trying to align whole genomes with MUSCLE is not going to be a very successful exercise.

[cliffbeall]

Perhaps use concatenated housekeeping genes?

e.g. here:

1. Science. 2006 Mar 3;311(5765):1283-7.

Toward automatic reconstruction of a highly resolved tree of life.

Ciccarelli FD, Doerks T, von Mering C, Creevey CJ, Snel B, Bork P.

PMID: 16513982

[lh3]

A bigger problem with muscle is that it not aware of rearrangement. It would be much better to use a proper whole-genome aligner such as Mauve. MUMmer is not ideal because it does not do multiple alignment.

[lwhitmore]

Does any one know how to output the percent identity for each alignment using the MUSCLE command line?