Phylogeny on 1800 protein sequences (alignment length 1500 characters)

[Michael Spitzer]

Hi!

I'm about to infer a tree on approx. 1800 aligned protein sequences,
with the lengt of the alignment being approx. 1500 characters.
I would like to ask here what kind of programs /algorithms are there
to get the tree as fast as possible.
Right now I'm using CLUSTALW's neighbor-joining with 100 bootstrap
trials.
The alignment procedure took ~18 hours on my 1,4GHz P3, but the
NJ/bootstrapping wasn't finished after ~5 days (job killed due to
power loss, it's my private PC...).

Are there any other NJ-implementations out there which are faster than
CLUSTAW and/or use parallelism (we have a 20 node SUN cluster (mixed
350 and 450 MHz CPUs...) at work, which (per node) is quite slow
compared to my 1,4Ghz P3, but utilizing more nodes could speed things
up)?
I tried PARBOOT (uses PHYLIP programs), but PHYLIP turned out to be a
lot slower than CLUSTALW (in fact the distance matrix calculation
step. A single CPU CLUSTALW job is nearly 20 times faster than ~20
node PARBOOT job on our cluster).

Thanks in advance,

Michael