The Workshop that Turns Researchers into Data Wizards
When people picture research breakthroughs, they likely imagine scientists peering at samples through a microscope and emerging with their Eureka(!) moment regarding a newfound discovery. In reality, it’s more like this:
Once a researcher designs and performs an experiment, the research process is far from over. Afterwards, there’s ample amounts of data to sift through. To find the answer to “Dr. Neeson’s” “Why,” he would probably work with a bioinformatician to analyze his dataset. These data scientists use data mining techniques to reveal the important takeaways of a researcher’s experiments. With that knowledge, researchers can better determine things like which signaling pathways in a tumor to target or the effectiveness of a particular drug. In turn, this knowledge fuels their future research decisions. However, the turnaround time can be lengthy, prolonging the wait for cures. Upwards of several months in some instances.
That’s where the Childhood Cancer Data Lab’s (CCDL) Data Science Training workshop comes in. At this hands-on, 3-day session held in Houston, researchers learned data science skills that could accelerate their own work. Drawing on skills learned at the workshop, childhood cancer researchers can perform basic analyses of their work to make informed decisions on how to proceed with their own research. Don’t just take our word for it, though. Read more and discover how the workshop’s incredibly valuable benefits through its attendees’ perspectives.
Better Questions = Better Results
Jim Amatruda, PhD, is no stranger to childhood cancer research. With several decades of experience, his knowledge of the field is first-rate. However, he leaped at the chance to learn something new with the CCDL data science training. He’s produced plenty of data, but often felt like he was left in the dark once it went to bioinformaticians.
“One of the problems is we pass the raw data off and wait for a result to come back, but you don’t have any sense of the process behind those results,” said Dr. Amatruda.
Already, the training gave him deeper insight into the particulars behind how bioinformaticians break down his data. Now, he can also gain added knowledge of his own data using skills, before he ships it off for deeper analysis. He believes his newfound skills could help every experiment be even more successful from the get-go.
“It’s an iterative cycle, you design the next experiment in a more informed way and get better data out of that,” said Dr. Amatruda. “The whole idea is to…gain insight from studies of tumors from our patients to really arrive at something that will make a difference and have it translated into a therapy.”
Dive Deeper Into Data
Starting to learn data analysis on her own seemed daunting to Collette. A research assistant in ALSF-funded researcher Jim Amatruda’s lab at University of Texas-Southwestern, she develops tumors in zebrafish to harvest and looks for potential targets in the cancerous cells that she could hit with novel treatments. With her programming experience limited to a semester in college, once the tumors were harvested, her expertise stopped. Now, she can dip her toes into preliminary data analysis thanks to the “starter kit” of code the CCDL developed.
“Rather than starting from a blank screen, which is a really scary prospect, this workshop gave me a package of code I can manipulate with my own data,” said Collette.
Instead of asking surface-level questions to her data science collaborators, she can dig deeper and request more robust analysis to accelerate her research.
“Now I can walk up to a colleague and tell them, ‘I’ve gotten to this point, what can we do for some next-level stuff?”
A Career-Changing Workshop
Lindsey felt frustrated. As a postdoctoral fellow at The University of Minnesota, she primarily relies on grants to sustain her work. However, her latest grant application was rejected when a reviewer claimed her concept — investigating gene expression differences between sexes — was too risky; there wasn’t enough data to back it up. Lindsey disagreed, she just didn’t have the skills to demonstrate it, nor the funds to pay a bioinformatician to do it for her.
So she started taking programming courses to learn those skills. Then, she discovered the CCDL’s data science training workshop, aimed specifically at childhood cancer researchers. Eureka!
Now, she feels comfortable plugging her pilot data into the code to demonstrate the validity of her research to a reviewer for funding using these valuable analysis tools. Three days of training and suddenly, funding feels much, much closer. “It seriously could be career-changing,” said Lindsey.
Cut Down on Wait Time
Imagine completing a huge project at work, sending it for review, and after months of waiting, getting a response saying the file wouldn’t open properly. What could’ve been an easy fix derailed the project for some time. Because many institutions only have a handful of bioinformaticians, researchers like Emmet can run into this problem quite often when sending data for analysis.
Emmet is a PhD candidate at Baylor College of Medicine whose lab developed a machine designed to help analyze individual cells in rapid succession. Initially, they merely wanted to ensure it worked, so they sent preliminary data to a bioinformatician. Instead of a quick yes or no, three months later he was told the samples weren’t usable. That’s a three month setback to their quest for cures. After attending the CCDL workshop, Emmet can check whether his samples provided quality data himself. Problem solved.
“Being able to run quality control on my own and know that the data I’m receiving is technically working is reassuring…if I can do my own basic analysis that would be great too. I’m not a bioinformatician yet, but this is a first step to understanding that.”
Clearly, these workshops are already teaching researchers new skills to accelerate their research to create safer, more effective treatments for kids with cancer. The results speak for themselves.The CCDL has two more workshops planned later this year in Chicago (June 24-26) and Philadelphia (October 14-16).