MUSCLE crashes

[Claudia Stein]

Hi,

I am new to phyloGenerator, and just following the Guide you provide I am always getting stuck because the program crashes with the error that MUSCLE cannot run. I am using a windows 7 machine. Any idea what I could do?

Thanks in advance,

C

[Will]

Hello,

Thanks for using the program and getting in touch.

In general, these sorts of problems seem to be caused by eccentricities of Windows that I'm not very good at understanding/diagnosing. There's an FAQ entry (http://willpearse.github.io/phyloGenerator/faq.html) that I've copy-pasted below that covers this particular issue. The short answer is that you might want to try downloading/running pG from your desktop. If you get the error with the example data, it's almost certainly this problem.

Another potential problem is that you haven't trimmed any huge sequences. Long sequences are flagged at the DNA Download prompt with "****" - depending on what you're doing, anything longer than ~2000 base pairs can cause a problem, and anything longer than ~10,000 base pairs will probably cause MUSCLE to crash. The short answer is that you should try using the 'trim' command. If you get the error only with your own data, it's almost certainly this problem.

Does that help in any way? If it doesn't, please do email me what you're putting into the program (e.g., species.txt, the genes you're searching for, etc.) and any other output you get (i.e., everything pG prints to screen) and I'll take a look. Every other case like this I've seen has been one of the above, so if it's not, I'd really like to know :D

Thanks again, and let me know how you get on,

Will

***************************************

Why is my directory not writable/why can't I write files to my working directory or to where phyloGenerator is installed?

For reasons that are unclear to me, Windows will let you install phyloGenerator to a place where it can't write out files (I say unclear to me, because you had to write files to copy phyloGenerator in the first place, right?...) This means pG can't pass information to other programs, because it can't write out the sequences it's downloaded for them to use. Picking an output working directory that you don't have permission to save files to will also cause problems! You should always be able to write files to your desktop, so if in doubt put phyloGenerator on your desktop and run it from there.That should fix this problem; I'm afraid I can't offer any advice about what parts of your comupter you have permission to write out files to because I'm not sat at your computer! Sorry!

---

Need a phylogeny? Try phyloGenerator

Measuring phylogenetic structure? Try install.packages('pez')

Want something to read? Try PEGE journal club or willeerd

Will Pearse
Post-doc, ecology / evolutionary biology
Davies and Peres-Neto labs
Skype: will.pearse
Cell: (+1) 514-973-1987

[Will]

Hello,

To bring the list up to speed on what happened off-list, once Claudia showed me her species list I was able to see that she was trying to build a phylogeny where she had very little sequence information.

You can't make a DNA alignment with only one sequence; there's nothing to compare all the sequences too. I should have written a check for this in pG I suppose; I might do so when I next push a version.

Thanks for bringing this to my attention Claudia!

Cheers,

Will

---

Need a phylogeny? Try phyloGenerator

Measuring phylogenetic structure? Try install.packages('pez')

Want something to read? Try PEGE journal club or willeerd

Will Pearse
Post-doc, ecology / evolutionary biology
Davies and Peres-Neto labs
Skype: will.pearse
Cell: (+1) 514-973-1987