Strange loops: journeys in declarative logic programming in genomics and beyond
Year: 2023
Type: article
Abstract: Declarative logic programming has long promised a more principled, expressive, and maintainable approach to computation, yet its adoption in domains like genomics and biomedicine has been uneven. In this talk, I present a series of “strange loops”—recurring attempts to apply declarative paradigms across bioinformatics, knowledge representation, and data integration—and reflect on both successes and failures along the way. Drawing from experiences spanning genome annotation, workflow systems, ontologies, and knowledge graphs, I explore how declarative approaches—from Prolog and Datalog to OWL and SPARQL—have been used to model biological knowledge, orchestrate complex analyses, and enable interoperable data ecosystems. While these approaches offer clear advantages in expressivity and abstraction, they often face resistance due to tooling gaps, usability challenges, and misalignment with existing engineering practices. I highlight key lessons learned: the importance of meeting users where they are, the critical role of standards and tooling, and the need to balance theoretical elegance with pragmatic utility. Finally, I discuss recent progress in declarative data modeling through frameworks like LinkML, and consider how emerging paradigms—particularly generative AI—may reshape the role of declarative systems as both specification and explanation.https://icfp23.sigplan.org/details/declmed-2023-papers/3/Strange-loops-journeys-in-declarative-logic-programming-in-genomics-and-beyondRecording: https://www.youtube.com/watch?v=CG24QFdVh-0
Author: Mungall, Christopher
Institution: Lawrence Berkeley National Laboratory
Language: English
Subfield: Information Systems and Management
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7154512751
