BEACON Researchers at Work: EDAMAME!

This week’s BEACON Researchers at Work blog post is by MSU faculty member Ashley Shade.

We received overwhelmingly positive feedback from our Explorations in Data Analysis for Metagenomic Advances in Microbial Ecology (EDAMAME) workshop last year, which was partially supported by BEACON and MSU’s iCER. It was immediately obvious that EDAMAME was addressing an urgent unmet need in the scientific community. The data from our course evaluation (thank you, BEACON, for facilitating our interactions with STEM ED, LLC to evaluate our workshop!) showed that EDAMAME learners had achieved our overarching goals of increased confidence and competence in microbial sequencing data analysis. In short, we seemed to be doing a lot right and having a positive impact on our learners’ abilities to own their microbial sequencing analyses. We were ecstatic.

EDAMAME Group photo credit: Tom Rayner, @tomonlocation

EDAMAME Group photo credit: Tom Rayner, @tomonlocation

This year, we had double the number of applicants from last year, and many of them were so exceptional that distinguishing among them for the purposes of admissions was challenging. We have reached out to recruit a broader applicant pool than last year, and I see the benefits of our efforts reflected in the diversity of backgrounds and academic interests represented among our EDAMAME learners this year. We also had a portion of our applicant pool from governmental organizations like the USGS and EPA, and also from not-for-profit organizations, which reflects a need beyond academia for the flavor of training that we provide. In retrospect, the high level of interest from many different universities, institutions and agencies makes sense: microbes are the functional foundations for all ecosystems, from human bodies to soils to deep-sea vents. Why shouldn’t there be wide interest in learning how to observe and analyze Earth’s ubiquitous but functionally elusive microbial communities?

At EDAMAME, we strive to provide the best learning materials and instruction that we can, and there is always room for improvement! For #edamame2015, we’ve refined our learning objectives (listed at the bottom of this post) and backwards-designed tutorials to meet those objectives. We’ve also spent more time on topics we glossed over last year like starting, using, and transferring files to an Amazon EC2 instance, which we hope will help learners who do not have access to a high performance computing cluster to know how to access the computing resources needed to execute analysis of our ever-larger sequencing datasets. After feedback for “MORE TIME” from last year’s EDAMAME learners, we expanded to 10 days so that we can spend more time with the more complex material. We also scheduled time for independent study with instructors available to give learners the opportunity to analyze their own datasets with our support. We additionally organized our tutorials in a GitHub wiki (with a CC-BY license) so that folks outside of the course can more easily find and use our materials. Help yourself!

TAs: Siobhan, Jackson, Paul, Sang-Hoon, and Jin

TAs: Siobhan, Jackson, Paul, Sang-Hoon, and Jin

And, this year, with almost a full year behind me on the tenure-track at Michigan State University, I also was able to bring with me my own new team of students and post-docs to serve as teaching assistants for the course. They bring contagious enthusiasm, patience, and experience to the course (they took EDAMAME in its inaugural year in 2014, even before joining my team). At every break, we collect “minute cards” (borrowed form Software Carpentry best practices) to receive immediate feedback from learners on what is going well and what needs to be addressed. With this feedback, our learners “spoke” loudly: our TAs were just amazing. I am so lucky to be supported by these stellar young scientists, including: Siobhan Cusack, Dr. Sang-Hoon Lee, and Jackson Sorensen from my group; Paul Wilburn from Elena Litchman’s group at Kellogg Biological Station; Dr. Jinlyung Choi from Adina Howe’s group at Iowa; and Aaron Garoutte from Jim Tiedje’s group at MSU.

We also recruited a celebrity line-up of local microbial ecology geniuses as guest speakers, including MSU’s Jim Tiedje, Matt Scholz, as well as RDP’s Jim Cole and Qiong Wang; KBS’s Sarah Evans and adjunct Ariane Peralta (East Carolina University); University of Michigan’s Vince Young, Vincent Denef, and members of the Schloss research team; Indiana University’s Jay Lennon; and the University of Notre Dame’s Stuart Jones.

Within a week and a half, our learners dive in to an array of computational and bioinformatics topics. We covered navigating the shell, cloud computing, remote sessions for running long jobs, within-sample and comparative diversity, merging paired end MiSeq reads and QIIME and mother for microbial amplicon analysis, shotgun metagenome analysis (assessing quality, digital normalization, assembly, annotation), R for ecological statistics, RDP tools and their new exciting targeted gene assembler Xander, using high performance computing resources, and accessing public databases. During breaks, there was volleyball and campfires and the backdrop of the summery Kellogg Biological Station on Gull Lake.

We were thankful and excited to be awarded transition funds from BEACON’s internal small grants competition to support our workshop this year until we found external funding. We couldn’t have continued the workshop without BEACON’s interim support. And… an announcement! I am pleased to share that the National Institutes of Health have taken an interest in EDAMAME, and we just have been awarded EDAMAME support for an additional three years! So, EDAMAME onward!

Thank you, again, BEACON for supporting EDAMAME!

Ashley Shade @ashley17061

Assistant Professor, Microbiology and Molecular Genetics

Newbie BEACON member – since 2014!

EDAMAME 2015 Learning Goals

  1. Increase computing literacy
  2. Develop proficiency in cloud computing
  3. Analyze microbial amplicon sequences
  4. Analyze microbial shotgun metagenome sequences
  5. Apply ecological statistics to analyze and interpret microbial sequencing data
  6. Access resources provided by public sequence databases

 

This entry was posted in Uncategorized. Bookmark the permalink.

Comments are closed.