| Characteristics: |
| Correspondence Analysis |
| Planar embedding |
| Interpretation |
| Storing annotations without freetext |
| Why? |
| How? |
| To what end? |
The signals for each gene are converted into numbers by some imaging
software and the numbers are stored. Although these numbers take the vast
majority of the storage space, they are simply-structured and therefore easy
to store.
But for biological interpretation of these numbers it is necessary to have detailed information about the investigated samples and experimental procedures (protocols) involved.
These experiment annotations are stored as well, mainly for three reasons:
- One likes to conserve these data over time. To still have it
after the experimenter is not available any more, maybe leaving a labjournal
that is not detailed enough or not readable at all :-).
- Also, in order to put the data into a larger context, that is to compare with
experiments done somewhere else. If one spins down yeast at 4 degrees or at
room temperature, it might well make a difference in transcription patterns
observed. But one will not be able to know why without access to these
parameters.
- Third, experiment description as well as sample biology are complex, resulting in too many experimental parameters to handle by visual inspection.
