Forum:Could Biostar Have Polls Or Rating Systems For Analysis Tools Or Databases

Question

[Ngsnewbie]

With the massive increase of next generation data, lots of new tools, R packages, databases and algorithms are coming out to play with them. To choose the right one for your analysis sometimes becomes confusing when you have a bunch of tools and not much time. Can we have a another section where users can create a poll (anybody can add options with time) and can give ratings to tools, databases or other things pertaining to bioinformatics analysis?

[Ngsnewbie]

I hope biostar will come up with some lucrative and relevant feature :)

Answer 1

[Istvan Albert]

The problem is that polls may not be all that well suited to characterize a complex data analysis process. I wish there was an easy answer to "what is the best tool to do X". The answer always depends on many factors: from problem specific ones to individual skills, abilities and preferences. After all that's the reason for having so many tools in the first place.

There is some value in polls if phrased as: What tools do you use in everyday work? etc. I will evaluate whether we could easily add that.

In general the best option is to check for answers to very similar questions - then ask for recommendations while also describing some specifics of the data that is to be analyzed while describing your expertise.

[Chris Miller]

Agreed. You need the right tool for the right job, and each tool makes tradeoffs. (performance vs accuracy, sensitivity vs specificity, etc) What's right for one experiment may be a dealbreaker for another.

[Michael]

Agree as well. A down vote could mean anything from I couldn't install it to "method inadequate" or implementation flaw. I suggest to use the Tools section to add real reviews of tools.

Answer 2

[dsbreak]

Maybe a better way to deal with the glut of tools is to have a place that lists the strengths of each tool. I find that obtaining this very high level summary takes a fair amount of time when searching google, seqanswers, biostars, pubmed, etc. While the wiki/Software Hub at seqanswers has a fair amount of information on a subset of tools, I still find it hard to quickly find a list of all the programs I should consider for say ChIP-Seq, let alone reasons to choose one over another.

Assuming anyone actually thinks this is a good idea, should this be hosted on biostars/seqanswers/a new site? I suppose it comes down to the capabilities of the hosting software, the community, etc.

[Istvan Albert]

My preference (and I will try to follow up on this with an example) is that of a tutorial that solves a problem with two-three different tools and while doing so comments/describes/demonstrates the various features that the tools have (or don't have).