I am no an expert in the use of R, JMP, SYSTAT and others. Although I have used them, I keep coming back to Excel. I just I am more productive with Excel. However, it has its limitations. Some graphical outputs are clumsy to produce in Excel (box and whisker plots, for example) and others are downright impossible (bee swarm plots). Hierarchical clustering, principal component analysis, and other types of analysis demand different software altogether. However, when it comes to sorting data, generating heat maps and bar charts, I find Excel works very well for me (except when it crashes).
If I was analyzing your data, this is how I would probably do it. For publication though, I would highly recommend any statistical treatments (T tests, ROC plots etc.) be repeated in a more “acceptable” statistical package.
The graphical outputs produced will include:
- Heat maps: the first step in any analysis. Helps to see immediately what is happening, to to flag any errors. and make some preliminary observations to guide subsequent analysis.
- F plots: the second most common type of plot I use. This is basically a comparison of the antibody profiles of two populations. It is represented by a clustered column or bar chart of the mean signal intensities of each group, which is overlaid with p-values.
- Receiver Operator Characteristic (ROC) Plots. This analysis allows you to determine how sensitive and specific a particular antigen (or group of antigens) is at discriminating between two populations, such as cases and controls.
- Box and whisker plots. blurb
Huw's Proteome Microarrays User Manual 