The whole world is in the midst of the COVID-19 chaos! I think that “social distancing” should not be a very painful pill to swallow for us. Most of the time, doing research is a “solitaire”. One of the unexpected benefits of “social distancing” is it leaves a surprisingly cleaner daily calendar. Without many out-of-control social obligations, the “distancing” actually leaves more precious time for each of us to spend meaningful time with students, friends, and family, who we care about the most.
The whole world is slowing down, ironically for most of us, in the midst of chaos. Is it “slowing down” a good thing? The slowing down allows us to think through things more carefully and design better experiments. It is a different way of working – a delayed, but “correct”, way to work more efficiently, which will benefit us in a long run. So, it is the time to embrace the slow pace (or you can call it less distraction). It is a journey forcing us to go back to the forgotten joy of uninterrupted deep thinking.
One of the biggest challenges for novice biomedical researchers, or even senior investigators alike, is how to navigate through the jungles of literature and design meaningful and smart experiment in their daily research life. It is hard indeed! Designing and redesigning an experiment requires a painful and seemingly never-ending literature search, comparing notes, and deep thinking – the process will mostly deliver frustration rather than instant gratification. Even more, to be a better researcher requires one to be skeptical and keep defying our ingenious ideas through the lens of “controls” and “alternatives hypothesis” – it is a painful process. Understandably, in the cycle of the researcher’s daily life, most of us tend to opt for more tangible “tasks” when we step into the lab. Surely, the gratification of finishing the task of the sub-culturing cell generates far more instant relief of an intense research day than staring at the computer screen, trying to sieve through the PubMed search results without generating an immediate checklist. As important as it is, frequently the “thinking” part of the day is often overtaken by the desire of finishing trivial tasks.
Is there a framework that could potentially help us to design a smarter experiment?
Asking the right question through a testable hypothesis
Aristotle’s theory of experimental science is hypothesis-driven. Despite the booming big data science advocating for a process of so-call data generating or hypothesis-generating, an experiment started from a clear goal (hypothesis) in the mind often leads to better experimental design. Restrain yourself from tackling too many ambiguous goals, and choose a more defined question. Regardless of how appealing or fancy the approach is, if your current best approach is not coupled with a very defined hypothesis, you will be very likely running a “garbage-in” and “garbage-out” experiment. Yes. I mean garbage data.
Clarify the purpose of a given experiment
To prevent running a “garbage” experiment, it is paramount to know the purpose of your experiment and what you will achieve in the best-case scenarios. Every experiment should have a purpose. The purpose can vary – from developing new methods to explore the provocative hypothesis. However, the purpose should be clearly perceived. Do this exercise: can you describe the goal of your new experiment in one sentence, NOW? If you hesitate for a second, think again!
Rethinking the true purpose and advantages of a complicated approach
The similar exercise of “clarifying the purpose” is also applicable to the approach you are going to take, especially the fancy and most likely expensive ones. One of the most exciting parts of discovery is to able to apply new technologies. The fast-evolving sequencing, imaging, and gene editing approaches are dazzling. However, craving for “cool” technology is a common sin when designing smart experiments. Adapting new protocols often takes time. Troubleshooting might lead to unexpected pitfuls which were not disclosed in hot technology papers. When attempting to take on new technology, be sure to wait a while until the technique is fully mature and the protocols have been reproduced by other labs. Be sure the new technology is absolutely essential to even explore your hypothesis and, importantly, it really can answer your hypothesis. If the easily interpretable western plot could answer your hypothesis, do it! And forget the phospho-proteomics. The golden rule is to seek reduction, simplification, and use of existing alternatives. Keep in mind, less is more.
Don’t fall in love with your hypothesis
The specific goal of different experiments might differ, but one thing is in common – the fundamental goal of designing an experiment is to test a null-hypothesis. In another word, to prove your favorite hypothesis wrong. When design experiments, we have to keep adding control control control in order to prove whatever we are going to see is not due to the technical or system noise/bias. The results are as good as controls. By default, we should spend most of our time thinking about controls. If the experimental group’s pattern is indeed different from the control, we are one small step closer to the truth. When interpreting the data, we have the keep pulling ourselves from our beloved hypothesis. There is a reason why the blind test is so desirable. Let the data speak! Yes. We love our hypothesis. Yet, intoxified by such love lead to the loss of sobriety. Carving for “as expected” positive results is the recipe for unintended wrong interpretation of data. If everything is simply as your expected, is it really a new discovery?
Respect others’ idea
Recent sociology research studied the key factors that lead to divorce. Guess that is the number one factor that leads to divorce? Contempt! Although it sounds seemingly not relevant to our research design, contempt oftentimes becomes an unconscious attitude when we “smart” scientists think about others’ ideas. Having some level of humility when discussing with peers, constructive brainstorming requires all of us to be critical of the goal and the design of the idea, but without contempt for the idea itself. Balancing mixed emotions while remaining confident is a heck of a skill to be polished every day.
