Last month, we challenged the Terra community to create workspaces showcasing reproducible analyses as part of the first ever Terra Open Science Contest. Four teams rose to the occasion and each submitted a great workspace crafted to maximize reproducibility (= can I run it), customizability (= can I tweak it) and documentation (= do I understand what's going on). It was tough to judge because all four were excellent in their own way. So we asked the community to help us judge these four entries, and now the results are in!
Grand Prize winner
We are delighted to announce that the winners of the Grand Prize are Kunaphas Kongkitimanon and Bhoom Suktitipat from Mahidol University in Bangkok, Thailand, for their workspace showcasing a somatic variant discovery analysis that integrates calls from GATK Mutect2 and Varscan2: https://app.terra.bio/#workspaces/terracontest/%20TOSC19-idap. Our community judges highlighted that their workspace was clearly documented and that the analysis was fully reproducible. We look forward to meeting Kunaphas at the BOSC conference in Basel later this month!
Our second place winner is Joshua Gould from the Golub lab at the Broad Institute, for his workspace showcasing Waddington-OT, a software package for analyzing snapshots of developmental processes: https://app.terra.bio/#workspaces/kco-tech/TOSC19-Waddington-OT. Joshua's workspace contains a set of nine tutorial notebooks that demonstrate how to use various features of the software, and is a great example of how a Terra workspace can serve as living documentation and self-service educational resource. Joshua will receive $5,000 worth of credits; we hope this will enable him to do more great work on the cloud.
In third place we have Denis Bauer and her team from CSIRO in Sydney, Australia, who delivered a workspace showcasing VariantSpark, a machine learning library for real-time genomic data analysis that uses Spark to scale up variant analysis to thousands of samples and millions of variants: https://app.terra.bio/#workspaces/fccredits-copper-blue-3695/TOSC19-VariantSpark. We were especially entertained by the premise of their notebook, which uses a synthetic HipsterIndex dataset to discover via GWAS what are the variants that predispose you to being a hipster. It's another great example of a ready-to-go tutorial workspace and earns Dr. Bauer's team $2,500 in credits to keep pushing the envelope of large-scale analyses on the cloud.
Finally, we’d like to thank Bo Li and Yiming Yang from Massachusetts General Hospital for contributing their workspace: https://app.terra.bio/#workspaces/fc-product-demo/TOSC19-scRNA-seq-cloud, which reproduces major steps in the analysis used to generate Figure 2b in Nature Communications, 2019. This is a great example of a workspace that supports the reproducibility of a published analysis, and it makes us wish we could award a fourth prize!
Please join us in congratulating all four teams and thanking them for their participation. We really appreciate the work they all put in, not only to participate in the contest, but also to help promote best practices for computational reproducibility and FAIR principles in biomedical research. Whenever more researchers adopt these practices, we all win.