RReportGenerator : Automatic reports from routine statistical analysis using R
Is something like this really useful?
RReportGenerator : Automatic reports from routine statistical analysis using R
W. Raffelsberger et al.
Bioinformatics Advance Access published online on November 24, 2007
With the establishment of high-throughput screening methods there is an increasing need for automatic analysis methods. Here we present RReportGenerator, a user-friendly portal for automatic routine analysis using the statistical platform R and Bioconductor. RReportGenerator is designed to analyze data using predefined analysis scenarios via a graphical user interface (GUI). A report in pdf-format combining text, figures and tables is automatically generated and results may be exported. To demonstrate suitable analysis tasks we provide direct web-access to a collection of analysis scenarios for summarizing data from transfected cell arrays (TCA), segmentation of CGH data, and microarray quality control and normalization.
I haven’t tried the package they describe, but it sounds like it is wrapping R for doing analysis from a GUI and then producing a PDF report from the results.
When I use R, I usually do not know exactly how to analyse my data, so it is always very exploratory and there is no way I could automate that. But then I am probably not the kind of user this package is aimed at, and I can certainly recognize the kind of R users that would be better off being sheltered from the gory details of R behind a GUI…
I don’t know, maybe I’ll try it out some time.
The citation for Research Blogger:
Raffelsberger, W., Krause, Y., Moulinier, L., Kieffer, D., Morand, A., Brino, L., Poch, O. (2007). RReportGenerator: automatic reports from routine statistical analysis using R. Bioinformatics, 24(2), 276-278. DOI: 10.1093/bioinformatics/btm556
January 25th, 2008 at 9:08 am
It’s true that this GUI doesn’t offer the flexibility you (may) need when you don’t know what you’re looking for in an explanatory statistical analysis. And this system is (still) far from intelligent automatic data analysis…
However, there are applications where a standardized routine way of doing some basic statistics can be very helpful in making yourself a quick idea about a new project. (Automatic) quality control of data that are typically produced the same way represents a classical application for such a system. And when you pass data to someone else, she/he might like to know that the data are actually of good quality (and a pdf documentation on this issue may be quite useful) …
Of course, some topics may be suited better for a routine automatic basic “analysis” than others. Anyway, this won’t replace in depth analysis by a (real) specialist who is able to consider properly all special features of the very data-set !
Simply, as a first step that goes all automatic, such an automatic analysis may save you quite some time and helps documenting the basic stuff, before you start your expert analysis.
One last point is that this system allows making some R functions and packages available for routine use to people who aren’t interested to invest in learning R syntax (and I’ve met quite a few of them).
January 26th, 2008 at 11:00 am
I am convinced that automated analysis is sensible for many applications and for many users. I think we can both agree, from what you write, that automated analysis can never replace detailed expert analysis, but it goes without saying that most types of data analysis does _not_ need expert analysis and that automated analysis suffices. For high throughput analysis it is the only option.
Is automated analysis needed, R is as sensible a language to implement it in as any. For statistical analysis it would also be my first choice.
Now whether I will personally gain anything from a package such as this is where I am in doubt. When I do data analysis it is rarely “routine” analysis so it is very hard to automate.
That being said, I could easily see myself adapt a tool like this for projects where I collaborate with biologist or similar who are less familiar with R.