Reasoning over taxonomic change: exploring alignments for the Perelleschus use case

Area of Science:

• Taxonomy and Systematics
• Computational Biology
• Bioinformatics

Background:

• Taxonomic classifications evolve with new data, leading to inconsistencies in name usage and tracking.
• Informal or outdated names can obscure phylogenetic relationships and taxonomic history.
• Accurate tracking of taxonomic perspectives is crucial for biological research.

Purpose of the Study:

• To develop and evaluate a novel approach for aligning biological taxonomies.
• To mitigate limitations caused by evolving classifications and inconsistent naming conventions.
• To enhance the logical consistency and expressiveness of taxonomic data.

Main Methods:

• Integration of human expert input with logic reasoners for taxonomy alignment.
• Utilizing the Perelleschus dataset with six historical taxonomies (1936-2013).
• Application of an Open Source reasoning toolkit to analyze taxonomic concept relations under various constraints.

Main Results:

• The novel approach successfully aligned 13 paired Perelleschus taxonomies.
• Reasoning process optimized logical consistency and inferred informative relations among taxonomic concepts.
• Information gain was, on average, one order of magnitude greater than in the input taxonomies.
• Merge visualizations effectively represented congruent and non-congruent taxonomic elements.

Conclusions:

• The human-expert and logic-reasoner approach offers a scalable solution for tracking taxonomic provenance.
• This method addresses differential biases in naming, phylogenetic resolution, and sampling.
• It provides a robust framework for managing evolving taxonomic perspectives and ensuring data integrity.