LifeGate: Extending a Global Visual PhyloMap into a Structured Biodiversity Knowledge System
Martin Freiberg, Birgitta König-Ries · From Mobilizing Data to AI-Ready Knowledge: Infrastructure for Multimodal Biodiversity Data
Summaries and key moments are machine-produced. Check the recording for the speaker’s exact words.
The short versionLifeGate's visual map of all described species will become an open, connected knowledge system with trait-based faceted search built from AI-assisted literature extraction.
What this was about
Martin (Freiberg) presented LifeGate's PhyloMap, a zoomable visual map of all ~2.6 million described species with ~5 million synonyms and published phylogenies since about 2000. Overlays expose gaps, for example GBIF coverage missing large parts of Coleoptera, algae and mosses, and IUCN assessments almost absent for arthropods. The goal is to attach traits so users can run a faceted search that narrows to one or two species. Birgitta König-Ries then described a new DFG-funded project in which biologists define identification traits, computer scientists use AI to extract them from the literature using manually annotated ground truth, and experts quality-check the results. It also plans to turn LifeGate from a standalone application into a connected part of the knowledge ecosystem through APIs, shared ontologies and knowledge graphs, and integration with Catalogue of Life and ChecklistBank.
Why it matters. Visualising all known species against aggregator coverage or Red List status makes data gaps obvious. A trait-based identification system built with other infrastructures could serve researchers, collections and the public.
In the room
- The PhyloMap covers all ~2.6 million described species, ~5 million synonyms and all published phylogenies since about 2000 (mostly molecular).
- Users can zoom from phylum to species and see phylogeny sources, synonyms, species counts and images.
- A GBIF overlay shows large gaps (e.g. Coleoptera, algae, mosses), and an IUCN Red List overlay shows arthropods essentially unassessed.
- Goal: attach traits for faceted search (e.g. leaves ≥15 cm, hairy, yellow corolla, in Australia) that narrows to one or two species.
- New DFG project: biologists define relevant traits from descriptions and existing trait ontologies; computer scientists with AI support extract them from literature using manually annotated ground truth; experts quality-control; newly detected species feed back in.
- Existing assets: website, code base, ~25 million photos from ~500,000 species, phylogenies and divergence times.
- Key gap: LifeGate is largely standalone. Plans are APIs, shared ontologies and knowledge graphs, reuse of others' extraction pipelines, integration with ChecklistBank and Catalogue of Life, and offering its visualisation to other projects such as collections.
Notable moments
Transcript
Automatically generated captions can contain mistakes, especially in names and technical terms. Times are relative to the room recording.
Read the transcript ↓
Related talks.
Experience of collaborative management of the World Auchenorrhyncha Database in TaxonWorks
TaxonWorks turned a personal taxonomic database into a collaborative, well-connected resource feeding Catalogue of Life, GBIF and iNaturalist.
Aligning An Evolving Global Data Model To Software With Global Ambitions: GBIF x TaxonWorks
TaxonWorks and DwC-DP have converged conceptually, but exchange between rich models will always be lossy, so mappings should be recorded explicitly.
BHL and the Planetary Knowledge Base: Structuring Biodiversity Literature for the Next Generation
NHM is re-OCRing all of BHL with modern models and extracting entities into a knowledge graph, with confidence gating to protect BHL data quality.