 |
|
In this practical, we have focused very heavily on quality metrics. However, there is still a lot of downstream analyses that we can perform.
Although, in this case, we are analysing sequences from a plant which has been fully sequenced, it is also possible to make microarrays de novo, to cDNA libraries. In these cases, microarray analysis would uncover a set of uncharacterised genes (with known sequences).
|
|
Another common task is to split genes into groups which behave similarly. This is normally the next step following a time-course experiment. For example, rather than asking which genes are over-expressed, which might also want to know about genes which are under-expressed following infection. Or genes which are are over-expressed at 6hr, but under at 12hr.
Clustering links together genes based on their expression profile — how their expression changes over time.
|
|
Finally, a common question is to ask "what do these genes of interest actually do"? An easy question to answer if you have 5 genes, but if you have 50? Is there any commonality between their functions?
|
|