TACC Supercomputers Power RNA-Seq Analysis Tools at Summer Bioinformatics Workshop

from HPC Wire by Makeda Easter,

Sept. 21 — Undergraduate biology labs are designed to prepare students for real-life biology work. These labs usually involve tried and true exercises like animal dissections, investigating enzymes, and microscope work.While traditional lab work is important, the field is rapidly evolving with the proliferation of big data and burgeoning technology. Computation and biology are now inextricably linked, perhaps signaling a need to trade in goggles for time manipulating the command line. But is the field ready for this change?

“The primary purpose is to get this into the classroom so that students will be doing the same experiments and working with the same datasets as any biologist in a lab at an institution,” said Jason Williams, Education, Outreach, and Training Lead at iPlant.

The iPlant Collaborative, a cyberinfrastructure project that gives researchers access to advanced computing, recently polled biologists across the country and found that 95% are working with large datasets. However, nearly two-thirds of researchers had little to no experience in bioinformatics, and only one-third said their institutions had adequate computational resources.To address this clear need, Dave Micklos, Director of the DNA Learning Center partnered with iPlant to develop a program that exposes undergraduate faculty to computational biology. The three-year, National Science Foundation (NSF) funded project, RNA-Seq for the Next Generation, arms faculty with the tools needed to teach bioinformatics to students.

“The primary purpose is to get this into the classroom so that students will be doing the same experiments and working with the same datasets as any biologist in a lab at an institution,” said Jason Williams, Education, Outreach, and Training Lead at iPlant.

The project is also ensuring equal access to high performance computing (HPC) resources and training by targeting public and minority serving institutions.

“Our goal is to reach faculty and students who want to learn how to do next generation sequencing but don’t have analysis tools at their fingertips,” said Mona Spector, staff molecular geneticist at the DNA Learning Center.

For the next generation

The project is centered on obtaining and analyzing next generation RNA sequencing (RNA-seq) data, which requires HPC computing resources. The technique gives researchers the ability to generate and analyze their own genome-scale datasets and answer novel research problems related to the transcriptome of any cell.

Examining RNA gives scientists a clear picture of what genes are being expressed and what is functionally relevant to the genome. But sequencing this information often generates terabytes of data that must be stored, processed, and analyzed to decipher meaning.

“The average bench biologist cannot analyze this data on their own,” said Williams. “Their options are they can either ask a collaborator to analyze the data for them or hire somebody to try and analyze it.”In the first workshop held summer of 2014, 11 faculty convened at Cold Spring Harbor Laboratory to learn RNA-seq techniques and brainstorm ways to integrate the technology in their classes. The programming was repeated this summer in two different cities for 33 faculty.

“The faculty expertise at the workshops was varied in regards to their knowledge about RNA sequencing and their computer skills,” said Spector. “This mirrored the challenges in formulating ideas of how to teach coursework to students with different knowledge levels as well.”

Led by Spector and Williams, participants learned how to analyze RNA-seq data using iPlant resources: Green Line of the DNA Subway and Discovery Environment which feature a simple interface that makes it easy for faculty and students to perform bioinformatics. Using the Agave API, the DNA Subway platform provides its users access to some of the most powerful supercomputers in the world for data analysis: Stampede and Lonestar at the Texas Advanced Computing Center (TACC).

“It was nice to be part of a group where we all do one technique and come together to develop teaching materials,” said Ray Enke, Assistant Professor of Biology, James Madison University.

Prior to the 2015 workshops, 26 faculty submitted 104 RNA samples which were sequenced at Cold Spring Harbor Laboratory’s Genome Center and the data were uploaded to the Data Store. Over the course of the weeklong training, faculty learned how to analyze this data. These projects were diverse and ranged from analyzing testicular gene expression patterns in infertile mice to examining Arabidopsis immune system changes.

In collaborative sessions, the groups also brainstormed ways to implement this technology into classroom lectures and labs.

“We trained biologists so that they would feel comfortable bringing their own RNA-seq experiments into the classroom,” said Williams. “We want researchers to combine their own interests with their teaching and use DNA Subway as a tool to not only analyze data for themselves, but to work with students as well.”

Impact: A look at two participants

While RNA sequencing is considered a gold standard of experimentation in his field, Ray Enke, a faculty member at James Madison University, had never actually done it himself. With “zero experience coding,” Enke was intimidated to become involved with the RNA-seq project, but soon discovered how much he could benefit from instruction and networking with other researchers.”It was nice to be part of a group where we all do one technique and come together to develop teaching materials,” said Enke. “Not many of my colleagues are using RNA sequencing so it’s nice to have that network.”

For Enke, the program has been beneficial for both his research on gene expression during vertebrate eye development and translating this knowledge into the classroom. Currently, he is working to further integrate the techniques he’s learned into his upper level Advanced Molecular Biology class.

“My ultimate goal is to start out with a cellular system, isolate RNA, perform RNA-seq, and do all of the analysis and validation on my own,” said Enke. “I’m taking baby steps to get there.”

Another participant from Hamline University, Irina Makarevitch, had extensive experience with sequencing but needed help implementing RNA-seq into the classroom.

“I applied to the program so I could develop teaching applications and exchange ideas with other faculty like me,” said Makarevitch.After returning from the workshop Makarevitch implemented what she learned over the summer. Using data analyzed on the Green Line on maize, Makarevitch developed guided inquiries to allow students to analyze data on their own. The activity taught students to ask independent questions, discover genes, and build graphs to interpret data.

“Students liked the idea that they were doing a real research project, as opposed to something where everyone knows the answers but are just going through the motions,” said Makarevitch.

Next year, the team will offer virtual training based on many videos that have been recorded from previous trainings. Williams and Spector anticipate a sort of multiplier effect, where faculty can share findings with peers and increase the number of students using these tools.

“Essentially, we’re building a resource where any teacher in the country can get training information and do these experiments themselves,” Williams said.

Source – HPC Wire

Leave a Reply

Your email address will not be published. Required fields are marked *


Time limit is exhausted. Please reload CAPTCHA.