Phylip's dnadist or paup* for general use

for syn and nsyn dists, you could try SNAP or HyPhy

Depends on the type of distances (mainly, which models) you want. EMBOSS' distmat is a good CLI program for a pipeline, I think, and does both uncorrected and corrected (by a choice from a few different models) distances.

http://emboss.sourceforge.net/apps/release/6.4/emboss/apps/distmat.html

Another possibility would be to use codeml from the PAML package (http://abacus.gene.ucl.ac.uk/software/paml.html) that allows you to estimate dN from amino-acid-constrained DNA sequences (i.e it uses a codon based model). If your pipeline is simple but takes long time or requires computational resources, you can try iPlant's discovery environment (http://www.iplantcollaborative.org/discover/discovery-environment)

I highly recommend raxMLGUI.

If you have the phylogenetic tree, CodeML from PAML is easy to implement in pipeline to recompute the branch lengths.

If you need to build the tree topology, you could use PhyML or RAxML.

I'd use DNAdist and Protdist, both availble from Phylip

http://evolution.genetics.washington.edu/phylip.html

Thanks to every body. I have used PALM successfully.