Comments on: TNT: a(n) (almost) comprehensive review

By: TNT: review of the graphical interface

TNT: review of the graphical interface — Thu, 08 May 2008 20:08:52 +0000

[...] Update: another review is available here. [...]

By: Paulo Nuin

Paulo Nuin — Tue, 26 Feb 2008 17:13:33 +0000

Hi
Indeed the Linux help is pretty poor, maybe because they devoted more effort to the Windows version with the GUI. It is quite difficult to understand what exactly can be done in Linux.

By: Francisco Collantes

Francisco Collantes — Tue, 26 Feb 2008 12:11:59 +0000

Hi
I know you cannot do it. I’m agree you that improving help is very necessary, at least for linux’s ursers. Also it’d be better if there were a manual (any kind of doc) with all commands.

By: Paulo Nuin

Paulo Nuin — Tue, 26 Feb 2008 00:34:01 +0000

Hi Francisco

I just did a review, I am not a developer of TNT so I can’t change it. It would be good to improve TNT’s help but there is little I can do.

Paulo

By: Francisco Collantes

Francisco Collantes — Mon, 25 Feb 2008 08:10:25 +0000

Thanks Paulo. I was so easy. I think I’d be a good idea to add your words in the includel html file in the package:

“unzip the package into a new directory and just run: ./tnt”

Now I’m going to start to “play” with the software.

Thanks again,
Paco

By: Paulo Nuin

Paulo Nuin — Fri, 22 Feb 2008 14:57:16 +0000

Hi Francisco

There is no installation procedure for Linux. The program is pre-compiled, so after you download it (http://www.zmuc.dk/public/phylogeny/tnt/tnt.tar.gz), unzip the package into a new directory and just run

./tnt

You should see TNT’s prompt then and you are ready to go.
Paulo

By: Francisco Collantes

Francisco Collantes — Fri, 22 Feb 2008 13:28:58 +0000

Could somebody explain how install in linux, please?

By: Paulo Nuin

Paulo Nuin — Fri, 01 Feb 2008 15:22:05 +0000

It is coming most probably Sunday or early Monday. I have a draft ready, including screen captures.

By: Matt

Matt — Fri, 01 Feb 2008 03:23:25 +0000

Meh- hard to argue with that! Looking forward to the review.

By: Paulo Nuin

Paulo Nuin — Fri, 01 Feb 2008 01:48:34 +0000

I agree with you. In part. I also believe the large problem will be the most sought in the coming years, but I still think the bulk of the phylogenetic analysis will be done with small sets.

Is TNT having only parsimony a “problem” for its popularity? I think so. Of course the fact that MrBayes only has bayesian is also a “problem”. Eclectic packages are the way to go for users, because most of the time you are not analysing the big problem, but rather a small one and if you have all methods in one place things get easier. I am writing an interface for a phylogenetic application that uses four command line programs and requires a couple of other applications to allow the user to complete the analysis.

TNT always had its niche and it is the best in it. It comes from a good heritage of programs and algorithms. Hennig86 was one of the best algorithms taking advantage of the limited (comparing to the present time) computer resources.

PAUP, in its eternal beta mode, also has drawbacks, and people that were very confortable using the GUI on a Mac now cannot use it anymore on the new Mac Intels. Phylip is also cryptic to use, the interface lacking user-friendliness.

I am reviewing TNT’s interface, you will read more in the coming days.

As anything in life, except for socks and some ball caps, there is no one-size-fits-all solution.