What General-Purpose Bioinformatics Platform Software Is Available?

Question

[Andrewjgrimm]

A colleague of mine is looking at setting up bioinformatics software for his Windows computer. He's planning to obtain a single platform that acts as an interface to all the applications he needs to use. Currently he has found Geneious, and he'd like to know if there's other platforms like it that he should also evaluate.

Some of the things he's after includes:

• Visual outputs: being able to display the structure of proteins visually, highlight motifs with color, rather than everything being text outputs.
• A menu based interface, rather than everything being command-line based.
• Containing the kitchen sink of different software, since he might be doing other analyses in the future, and he'd rather not install new applications for each analysis.
• To a reasonable extent, a common interface for all of the applications the platform utilizes.
• Ideally: user help that doesn't just describe the switches for each application, but describes what kind of application you'd use for a certain scenario.
• It's ok for the platform to be commercial/proprietary.

[Casbon]

Anyone tried http://www.bioclipse.net/ ?

Answer 1

[Neilfws]

Given the needs that you describe, both Geneious and CLCBio sound like good options. I'd like to raise a couple of points (and I realise that they are not all relevant to this situation).

First, the tools that you require depend on precisely what it is that you want to do. A "kitchen sink" of software, covering everything that you may potentially want to do, including things that you have not thought of yet, simply does not exist. This is one reason why bioinformaticians prefer open-source solutions and platforms such as Linux. They provide a toolkit, made from parts that do small tasks very well, which can be mixed and matched as required.

Second, I would consider web-based tools. In the past, one reason for installing applications locally (on any platform) was that web-based tools were not very good. This situation is changing, with many institutions providing web interfaces to powerful tools - a good example is Galaxy. Now, this does not address the issue of a common interface but if you think of "the web" as "one place" - as a platform - and you have a good system for bookmarking and annotating the tools that you find on the web, you can go a long way towards avoiding local installation. Frankly, if someone is desperate to avoid text output, command lines, installation and having to read documentation, I'm not sure that any kind of local computation is for them :-)

[Will]

+1 for Galaxy ... I've made a local fork and added all of my analysis tools. This makes reproducibility and "tool transfer" between projects a cinch.

[Elipapa]

@Will I am curious to know how do you 'share' a local fork of Galaxy (say on your laptop) with other collaborators? I can't think of a practical solution

Answer 2

[Pierre Lindenbaum]

NCBI genome workbench

Answer 3

[Khader Shameer]

Bioinformatics Work bench Solution(s) from CLCBio will be a good option.

Even though it may not be exhaustive, I would recommend to try open source alternatives like Jemboss or Mobyle. Earlier version of Mobyle was available as a local installation, not sure if it is possible with current version.

Answer 4

[Bioinfo]

I would recommend the CLCBio solutions. It has solid visualization capabilities, and the wizard driven interface, along with detailed help make it a tool with low activation energy to get going quickly with actual computations. The platform also has the ability to be a common interface to open source and in house algorithms. They provide an SDK so users can develop (or in this case probably pay CLC or another to develop) customized workflows or plug-in. There is a server based enterprise architecture that will allow you to scale. The server has some added capabilities. There is a “3rd party plugin tool” which allows you to reference 3rd party tools from the command line (such a bowtie or samtools). This interface is constructed through a web based thing client GUI.

I have evaluated several options such as Genious, as well as cloud based offering and found that CLC provides a solid baseline infrastructure, but also the ability to extend this as needed. As to cloud based solutions, this can be a great solutions. Genome quest and DNANexus have compelling products, but neither right now has an API to customize things much (both have plans for this though), and if you are moving lots of data to the cloud, you better have the appropriate internet pipes. As to open source vs. commercial, one thing often overlooked is the maintenance and support costs – in my interaction with CLC they have knowledgeable scientists and consulting teams that stand behind their product and respond promptly to issues.

And no I don’t work for CLC ;-) We do however use CLC enterprise as a baseline infrastructure - along with open source and in house SW and algorithms - to build out our pipelines to support 6 NGS machines running 24/7 and provide our scientists the ability to visualize and add value to the data we generate and analyze for them.