One thing you won't learn at school is all the complications involved in validating a target. The best resource I know of for this is the "in the pipeline" blog by Derek Lowe. I have spent 100s of hours reading all his posts (they go back 10 years) and especially the comments. It is a hard slog, but the amount you will learn is staggering.
Edit. I guess is should include the link http://pipeline.corante.com/
Things appear hard and severely constrained because we habitually fall into the trap of thinking linearly and incrementally, as opposed to, at least sometimes, multi-linearly and in bounds, by perspective shifting. To do that one has to question the fundamental assumptions of which this linear trajectory were founded on. More often than not, one finds that much of what is considered so is merely based on someone', usually a group's very convincing, albeit not necessarily correct narrative.
As wiki puts it:
http://en.wikipedia.org/wiki/Drug_discovery#Drug_targets
Generally, the "target" is the naturally existing cellular or molecular structure involved in the pathology of interest that the drug-in-development is meant to act on.
[it's] low-input, high-throughput, no-output biology