Dear John,

Something has indeed gone wrong with the server. I restarted job queues, so your analyses should have finished. I am not in San Diego at the moment, but when I get back, I'll check to see what caused the problem. In the meantime, the cluster should run fine.

There is no way to cancel a job from the web (once it goes in the queue, it cannot be removed). I might change that later, but it's a bit tedious to implement (because you don't want people to be able to cancel others' jobs essentially).

Thanks for bringing my attention to the problem.

Sergei

Dear Amanda,

If your alignment has more than 25 sequences you won't be given the option to run REL (due to our need to limit job-run time; a REL analysis with many more sequences can tie up the cluster for several hours). Similarly, FEL analyses are available for alignments with <=50 sequences.

You can always execute REL-like analyses by downloading HyPhy and running them on your own computer (Standard Analyses->CodonRateAnalyses->dNdSRateAnalysis.bf).

HTH,
Sergei