Day 96: The art of requesting data
I have some data. This shouldn’t be news, I’ve been talking about how I’m trying to train a model to predict an output using this data and all the woes that have come from the training process. Turns out I have a problem, well more like questions that need answers and for that I need help.
Here’s the deal, I need data. Not just any data (help! sorry I couldn’t … help myself). So as I’ve said, I have some novel biological data that I’ve collected (yay). To verify that this data is what I say it is, I need something to compare it to. Now, that is a problem since this data is new, hence novel. Essentially this means that there are no comparable data sets that I can use, it’s all very frustrating.
Fear not! My data is human collected, so because there are no human data sets that are comparable, I need to look elsewhere. As with any good research question in humans, there is more than likely an animal model that exists. In my case I found just that! This is great news and if everything had worked out perfectly I might not be typing this…. however.
The studies I’ve found with data collected from mice were published back in 2013-2014 practically a lifetime ago in the academic world. There were several follow up studies, however they went in a different direction with the experiments, so therefore the only data set that I could even remotely find that would let me draw a comparison with my data is about six years old.
Now comes the fun bit, requesting data from a potential rival. That’s the problem in research, there is a fear of being out scooped. This fear makes it hard to get data that has already been collected via another lab. Because I don’t have the time or resources to collect the same mouse data, I really need to hope they will be willing to share.
“In academica we live by one simple rule, publish or perish.”
Keep in mind they have no legal reason to share (at least in this case), just a (possibly) moral one. They could tell me that because my research is similar to theirs, they cannot risk sharing the data and having me find something that they missed. This makes sense in a roundabout way because in academica we live by one simple rule, publish or perish. It’s not that they would look bad if I found something interesting, they would just miss the chance to find it themselves, even with data that is six years old.
With that in mind, I sent a cold email (the worst kind of email) to the corresponding author and held my breath hoping that not only would they be willing to share the data set, but that the data set still existed. Would you believe they responded? Not only did they respond, they did so within the hour. That is no small feat in the academic setting.
I was told by the first author, a grad student at the time and now a full professor, that after several moves and setting up their own lab, they had no idea where the data was. I wasn’t expecting a response, but if I had, that is the response I would’ve expected. They did give me hope though. They suggested that I reach out to the head of the lab the work was performed in. They might still have the data readily available for me.
For those of you keeping score, this is now cold email number two. This time to someone who has been doing this far longer (read: less likely to share the data for the reasons I just mentioned). Two days ago I sent the email with fingers crossed and I got a response. Again, if I had expected a response, it would be the one I got.
They wanted me to explain who I am, what I am researching, who I am working under, and lastly, what I would be using the data for. Publish or perish is very much a real fear. This is where you have to rely on something you were never taught explicitly in academia. You have to rely on diplomacy, kindness, and pure luck.
So yesterday I crafted my response. I explained that I am a fellow researcher in a field adjacent to there own. I tried to paint a picture that was non-threatening. I explained why I needed the data and what I would use it for. That I am very early in my research and they have nothing to fear, I wouldn’t use their data to do something they could do. I would use it, yes, but within a context that they wouldn’t. There wouldn’t be a publication or findings that would make us rivals. I was a friend. I just needed to verify my own work and I needed their help.
It’s now the weekend, so I’m not expecting a response until monday at the earliest, if I even hear back. Data requests are difficult, I’ve seen colleagues trying to do very nobel things be turned down explicitly because they are the competition. I’ve seen colleagues turned down because there was fear that they would find something or do something with the data that the original lab didn’t think of doing or couldn’t do.
Now I wait to see if that fear keeps me from the data I desperately need to verify my findings. Can I do it without the data, yes of course. There are other ways and more experiments that I can do. However, without the data my work will be slowed, which admittedly will benefit the competition, but at the cost of helping people.
I don’t know what will come next in this story, but I do know that while hoarding data may help the lab that has the data, it hurts research in general. We are all in this together and maybe they were harmless questions. Maybe the PI of the lab just wanted to make sure that I was actually going to use the data so they didn’t waste their time looking for it only to have it not used. Maybe fear wasn’t part of the equation. However, if I was the kind of person that would make a bet, it would be that the fear is real and despite my best efforts to show that I mean no harm, that fear may keep me from the data I need.
My lovely readers, I implore you, should you find yourself in possession of data that is being requested, don’t be afraid. It’s time we share, even at the risk of losing out on a finding that you may not have ever thought to look for. A scary prospect I’m sure, but when we are talking about medical research in particular, the consequence is potential harm to some of the most vulnerable of us.
I promise, should I decide to become a PI of my own lab, if I have data you need I won’t let fear keep me from sharing. As scary as it is, we’re all in this together. Let’s stand on the shoulders of each other and accomplish something more than any one of us could alone.
Until next time, don’t stop learning!