We just had an overview presentation of emboss (The European Molecular Biology Open Software Suite) this week. It seems to be packed full of interesting programs, but I wonder if for most of those there are not already many interesting alternatives.
I would like to have your opinion about whether or not this particular suite is worth investing some time (let's say the equivalent of 2 working days) to learn. What are in your opinion the pros and cons of this suite or whether one should learn to use something else.
I would say yes, it is worth learning the various file format conventions and what tools are available in the suite. There are a lot of gems in there, that you end up coming around to since flashier alternatives are not maintained, don't run cleanly, don't work as advertised, etc. For example, I find dottup to be the best command line dot-plot program for producing images (clearly showing my age here, I know). Additionally, since Galaxy wraps the EMBOSS suite, you have the power of Galaxy workflows and reproducibility at your fingertips for all the tools in EMBOSS, which are not currently available for many alternatives.
EDIT: It looks also as is many of the tools in the EMBOSS suite are available as web-services from EBI's SoapLab which can be utilized by Taverna.
+1 For Casey (nice you like it already at such a young age ;-) ). It is indeed great that it is in Galaxy. In fact it is also part of NEBC Bio-Linux (which we happen to use on NuGO's blackbox network).
Importantly EMBOSS is open source. So if there really are parts that you don't like that much you can improve them and keep the rest.
The good thing about these old tools is that they have been debugged and optimized for ages. In this sense they are comparable to the basic Unix tools, which often perform amazingly well for their age, or actually because of their age.
If this presentation is for the starter then I believe it's worthy spending two days time. Moreover, emboss is a gem tool covering the different aspect of sequence analysis ranging from alignment-phylogeny and workflow management.EMBOSS integrating a wide range of currently available packages and tools for sequence analysis into a seamless whole makes a great package. I did course on sequence analayis with Embster which uses emboss too and can't find the boundry of this tool. So why not emboss?
EMBOSS always reminds me of the linux/unix command line. Many simple (as opposed to overly complex) tools that are easy to use and get their job done, readily available at your finger tips. It's up to you to combine them into a powerful pipeline according to your specific needs. If you do work with sequence data, it is definitely worth spending time to learn it. It will pay in the long run.
When looking at Biostar I also find a lot of evidence for how useful EMBOSS tools are. I remember a bunch of questions from the last weeks, that could be solved using EMBOSS tools, e.g. questions about ORF finding (getORF) and translation (transeq, sixpack), primer prediction (primer3), etc.
I would like to stress the very good code maturity, too, an aspect that might often be under-evaluated when comparing to more novel algorithms or implementations.
Though, I would not consider them as a resource for the most advanced novel algorithms, but more like the reliable work-horse for everyday analysis (for example, I think, the phylogeny part is based on PHYLIP, which is not being seen as near state-of-the-art, I remember from a question here).
To learn these tools I would maybe try to focus on the different tasks they can be used for and look at a few tools in depth, such that when you want to solve a problem, you will remember to look in emboss tools for an application.