Valley of Ideas: The Importance of Review Articles

I wanted to pen (type) some thoughts down on what the initial portion of the process of coming up with a research idea is like.

I can sum it up in one phrase: it looks like a valley. And there’s nothing to be ashamed of having difficulties crossing through it.

I came into the influenza world as a total greenhorn. I came into the machine learning world on the back of some self-taught Python. Math and biology were my worst high school subjects, and I barely scraped a B- on my virology class in undergrad. Lo and behold, here I am doing virology, more specifically, mathematical modelling of flu transmission using network analysis methods and data-sciencey techniques.

So when a greenhorn like me comes into the flu world, I’m looking at the field and going, “WOW THERE’S SO MANY PROBLEMS THAT CAN BE TACKLED!”

I ended up choosing the problem of virus reassortment as the central theme, and started formulating some initial thoughts to the sub-problems that I thought I could give a go at.

At this stage, we’re looking at the problem with a naive pair of eyes. We’re thinking, “How hard could it get?” That’s like standing at one side of the valley. The other side looks not too far off, the distance looks “trekkable” without much issue, and the pot of gold waiting at the other side of the valley looks so promising.

If we’ve been trained right, either by hook or by crook, by one-to-one coaching or by fumbling our way through, we begin digging through the literature. We don’t do experiments first. We start familiarizing ourselves with what other people have done. The problem is, if we started out naively, then we don’t have a GPS device on hand. All we’ve got is a flashlight. As we descend into the valley, thorn bushes, trees, and fog conspire to keep us from keeping our eyes on the other side of the valley. As we descend into deeper depths of specialized knowledge, we start losing focus of the bigger question.

That’s when we’re really in the thick of the battle. You could call this a mini version of Uri Alon’s “cloud”. This is not the “cloud” as he understands it in its entirety, but this is the beginning part of the “cloud”. And it’s already damn foggy out there.

What happens next?

Well, at some point, we find a map laying around, which looks oddly familiar. IT’S THE MAP IN OUR BAG, which we forgot to take out. And yes, there’s a compass right next to it.

IT’S THE REVIEW ARTICLES, DAMNIT! “Why didn’t you take out the map at the beginning?” Well, crap like forgetting to take out your map happens. Therefore, that same crappy feeling that comes when we realize we should have read that damn review article better also happens. We move past the self-imposed shame (for stupidity and forgetfulness), and move on to scrutinize the map carefully.

But the map is incomplete. The map laid out only what major landmarks were spotted along the journey to the elusive other side of the valley. It didn’t lay out the exact path to the other valley. We sigh a sigh of relief – if someone else had mapped out the journey to the other side, it’d be a boring journey. They’d look back retrospectively, write the map that charted the path, and moved on in search of the next valley and its pot of gold. But no, nobody’s gotten there just yet.

With the map in hand, though, we see the rough location of where we are now, and where the furthest paths have been laid down. Researchers have tried different approaches, and they all seem to be converging on one potential path that will lead up the hill.

We pick up our map, our hiking gear, and move on. There is much work to be done.

If there’s been one big lesson I’ve learned from the past two months of trying to figure out how influenza packaging can affect reassortment, then that lesson has been to read the review papers. And to not fall asleep while doing that. Review articles lay out the landscape of what’s been done so far, and some propose further directions on what needs to be done. Others may not.

In my case, carefully taking the time to read the review article could have saved me a ton of time (by my own tracking, at least a week). Instead of being frustrated over the incongruence of some ideas in my mind with what was being published, I could have used the time to crystallize (through writing down) my thoughts and to put, in writing, the direction I thought would be best going forward, and the reasons for that.

That said, I’ve made a conscious choice not to bash myself up over that. Rather, I’ve decided to pen it down and share it out. Hopefully, you’ve found it useful. And that you’d READ THE DAMN REVIEW ARTICLE! 🙂