There are exceptions to every rule. But here are some guidelines to help you figure out what I am asking for, when I ask for a good hypothesis on a topic:
1. interesting – it should be something worth finding out, not obviously true or false
2. not already known, but sensible - e.g. not "vitamin C prevents scurvy", we already know that. But based on what you know, it should make pretty good sense. “the maximum growth rate of the hybridoma will be at pH 7.1” is too specific and not really interesting – no basis, but “a small change in pH of the medium will have a large impact on the growth rate” is sensible and interesting.
3. experimentally falsifiable - it is a statement, not a question, not "how much vitamin C should a person take?" but "at a certain dosage, excess vitamin C will be detected in the urine"
4. feasible to test - a good hypothesis will be something that can be tested with facilities available for a reasonable cost - e.g. not "If two groups of people are raised their entire lives on different dosages of vitamin C (60 vs. 500 mg) but everything else identical, then the group taking the higher dose will be healthier", because that’s far too long and expensive to test. Generally, undergraduates should avoid human experimentation, certainly any biochemical, molecular biology research is unsuitable for a novice.
5. simple enough that it can be tested - "Given 500 mg of vitamin C daily, a group of smokers will excrete less vitamin C in their urine than a group of non-smokers". If the protocols you suggest are not right out of the lab manual, you had better be sure they are sensible, clear, and possible, and that you give enough evidence in your answer that I believe those things. Don’t say, “find the gene, put it in a mouse, and see if it works.”
HOW TO FIND A GOOD HYPOTHESIS?
* Start with an interesting idea.
Try to define how you would go about testing the idea. Usually, the first or second step is too general or not clear. Take the hypothesis above, “find the gene, put it in a mouse, and see if it works.” Which gene? Well, pick one. The sickle cell anemia gene, for instance.
* So now I have, “find the sickle cell anemia gene, put it in a mouse, see if it works”. I can find the human sickle cell anemia gene on the world wide web. Will it work in a mouse? Simpler first step hypothesis: “Put the sickle cell anemia gene of a human in a mouse and see if it works”. Oops, mice already have hemoglobin genes. So I have to substitute one for the other. So I have to find a way to substitute. To do that, first I have to be able to target a particular sequence, where the gene is.
* Hypothesis “I can design a specific agent to mutate a particular spot in the mouse genome.” Too vague. Let’s try just mouse cells to start with. And try a sensible mutating agent. “A DNA fragment with a mutation in the hemoglobin gene of a mouse will substitute for the actual gene if I add it to tissue culture of mouse cells.”