This question belongs in "Cross-Validated" as well as "Biology"
I am looking for example datasets that I could use both for teaching and for my personal training, that would be good examples of very successful applications of in silico data analysis tools.
I am interested in tools like: k-means or hierachical clustering, principal component analysis, generalized linear models, or basic machine learning like class discovery & prediction.
Of course, if the data is available, that would be even better, but a simple figure in a paper would work as well.
What I mean by good is "fairly unambiguous and demonstrative", ie. the results could be, for example, a dichotomy of the dataset into two or three groups that make sense from a biological point of view (e.g. different tissues of the same organ, different times in cell development...).
A good example would be, for example, this figure:
Rows are genes (microarray expression analysis) and columns are healthy (HV) or cancer patients (AML); clustering makes two large clusters appear.
It is a really good example because we can easily check that the two clusters match the two types of patient samples, which makes this clustering a good, pedagogic case study.
In other words I am looking for striking examples of how the visualized data makes perfect sense, if properly analyzed.