Tonight I started watching the ISME19 Workshop by KBase: “From Reads to Function, Day 1.” Mikayla Borton was the first speaker of this session. The workshop had twenty-five people in person and ~220 virtual participants… from fifty countries. The KBase Education team was there! Borton’s session was on the Genome Resolved Open Watersheds database (GROWdb) project and data. This was an effort to sequence/sample global rivers. To obtain a large sample set, the effort was a community science initiative. As rivers flow from headwaters and circulate through populations, their composition changes. The first phase recovered more than 3000 genomes from US rivers. The sample set for the workshop was metagenomes from the Congo river. Byron Crump from Oregon State University spoke about the samples. Crump noted that this is one of the largest river basins globally. They took samples and studied the biogeochemistry of carbon. Crump studied samples to calculate travel times and the average age of the water sample. The team asked: does landscape and hydrology structure river microbial communities? Ted Bambakidis used 16S sequencing to analyze samples. They found that the travel time influenced the microbial composition in large rivers. In small rivers, microbiomes were still affected by travel time but not as much. Microbiomes were structured by wetlands, pH, and dissolved oxygen content. These results helped Crump develop the “Microbiomes and the River Continuum Concept” in which terrestrial microbiomes and planktonic microbiomes influence community structure.
Next, Benjamin Allen from Oak Ridge National Laboratory gave a KBase workshop. They defined KBase as “a community-driven research platform for systems biology.” Allen spoke about creating and managing FAIR data and then analyzing it with collaborators on KBase. Allen described the components of KBase narratives: toolbar, data, and apps windows as well as markdown, code, app, and data object viewer/cells. The KBase narrative is built on Jupyter notebooks. Allen provided an overview of customizable analysis possible with KBase: genome assembly and annotation, comparative genomics, metabolic modeling, transcriptomics, viromics, and metagenomics… and interconnections! Allen explained that the compute is provided by the Department of Energy. KBase also has robust public data integration with RefSeq, JGI plant and fungal genomes. Narratives can be kept private or shared with collaborators or the KBase public. Allen shared that a new KBase feature is “Collections.” This new feature allows exploration of datasets in a “dashboard” format. Allen also spoke about modeling across systems and scales by creating data structures and analytical tools. The KBase team has about forty staff members.
