Jason, that's very interesting.

I advice you to use TNT. I also advice you to carfully consider how you code your characters.

http://www.cladistics.com/aboutTNT.html

Maybe you should take a look at Simmap:

http://www.simmap.com/index.html

Leandro, thanks for the recommendation. Do you know if there is any tutorial for TNT? I don't a strong background on coding.

I recommended to you Simmap, but there is also MEGA which has a simple graphical interface. 

http://www.megasoftware.net

Maybe mesquite could do the job too:

https://mesquiteproject.wikispaces.com/home

Yes; please check the TNT download page: http://www.lillo.org.ar/phylogeny/tnt/

Unfortunately, Simmap is only for MAC.

I am still having a hard time to grasp TNT? Any other advice?

Mega works on all operating systems.

Does Mega work for morphological data? Is there a tutorial out there that can walk me from beginning (creating matrix) to the end (making a tree) 

Thank you Antony

I think Mega can handle morphological data (note sure).

http://www.megasoftware.net/tutorial.php

But I know mesquite can.

Dear Jason,

To be honest, I do not see why you would want to use a phylogenetic program for this. Phenetic clustering would seem more appropriate for the type of data, and possibly the type of questions you have. SPSS will provide all you need.

I agree with Marc. I am not sure whether the processes of tumorigenesis are similar to cladogenesis. So it may be not appropriate to apply a phylogenetic analysis in your case. However, you may try UPGMA and PCoA both of which are based on similarity distance of characters.

Dear Jason, 

It is not a case of phylogenetic analysis. You should use a cluster analisis like UPGMA or WPGMA. A good program could be PAST, NTSYS, SPSS, R... with Bray-Curtis or Euclidian Distance Index. To work with a siminality matrix, you might use MANOVA or PERMANOVA analysis. I Suggest you PERMANOVA.  

You might want to check these refereces out:

http://www.ncbi.nlm.nih.gov/pubmed/?term=tumor+phylogenetics

Dear All,

You guys are awesome. Thank you so much for contributing to my question. I will definitely update you guys once we figure it out!

Jason

yes I  agree with the multivariate analysis , clustering and principal component analysis