Event
QLS Seminar Series - Matthew Akamatsu
Tuesday, March 24, 2026 12:00to13:00
Extensible and modular research collaboration with lab discourse graphs
Matthew Akamatsu, University of Washington
Tuesday March 24, 12-1pm
Zoom Link:听
In Person: 550 Sherbrooke, Room 189
Abstract:听In their projects, researchers face the challenge of getting up to speed on the state of knowledge in their research area, tracking their results, and sharing key results with their lab and collaborators. To address these barriers, we created discourse graphs, a system and application for modular research collaboration. Discourse graphs decouple scientific research into its atomic components - questions, claims, and evidence - and connect them into an extensible graph. This allows researchers to inventory discrete observations from the literature and their ongoing research, and synthesize them into evolving scientific stories.
Our cell biology lab at the University of Washington shares a lab discourse graph as a graph-based lab notebook, project tracker, meetings record, literature parser, and scientific story compilation board. An ongoing pilot with 10 labs has demonstrated that our open-source plugins for popular notetaking apps help researchers think like a scientist, remain oriented to their target question/hypothesis, and make modular contributions to shared research projects. This scaffolded approach to scientific projects lowers the barrier to meaningful research contribution by new researchers.
