Novel Euler/X diagnostics for inconsistent input alignments.
Two taxonomies together with 9 input articulations are shown in the TAP (taxonomy alignment problem) on the left. It is not clear what the “combined taxonomy” looks like. For this example, calling Euler/X’s “check-consistency” command yields “No” – the TAP is inconsistent, i.e., self-contradictory. Even for the trained expert it is difficult to locate the error. Euler/X can execute a “black-box inconsistency analysis”, trying various combinations of articulations to see which subsets of articulations are consistent. The resulting diagnosis lattice has 2^9 = 512 nodes, and is therefore hard to navigate. Euler/X can extract from the 512 combinations 9 maximal consistent sets (i.e., the purple combinations in the lower right cannot be extended with any of the missing articulations without rendering the alignment inconsistent), and 2 minimal inconsistent sets (i.e., the red subsets are already inconsistent, but removing any articulation will yield a consistent subset). This allows the user to effectively explore and understand the consistency/inconsistency landscape, leading to efficient repair actions.
Euler/X toolkit visualization of Maximal Consistent Sets (MCS) and Minimal Inconsistent Sets (MIS).
This 8-minute screenshot video, produced in late April, 2015, demonstrates the interaction of the Exploring Taxonomic Concepts (ETC) and Euler/X software tools.
The ETC team has built a highly useful and effective front-end web service to the Euler/X reasoning services, allowing users to import two concept taxonomies through an on-line file management system, provide articulations, run input visualizations, obtain logically consistent alignments, and extract these products from the web interface to the desktop. The underlying ideas and methods are explained in Franz et al. (2015).
The capabilities of the user interface are illustrated with the Minyomerus use case, aligning two classifications authored in 2015 and 1982, respectively.
ETC Euler Taxonomic Concept Alignment Demonstration – Minyomerus 2015 versus 1982 from taxonbytes on Vimeo.
A short paper related to the Euler/X toolkit and concept taxonomy alignment project has been published. It deals with the issue of diagnosing inconsistent input constraints in an attempted pairwise taxonomy alignment, analyzing and visualizing their logical provenance so that the user can localize the inconsistencies and proceed towards repairing them. These logic services are already implemented in the toolkit.
Abstract. Derivations and proofs are a form of provenance in automated deduction that can assist users in understanding how reasoners derive logical consequences from premises. However, system-generated proofs are often overly complex or detailed, and making sense of them is non-trivial. Conversely, without any form of provenance, it is just as hard to know why a certain fact was derived. We study provenance in the application of Euler/X, a logic-based toolkit for aligning multiple biological taxonomies. We propose a combination of approaches to explain both, logical inconsistencies in the input alignment, and the derivation of new facts in the output taxonomies.
Chen, M., S. Yu, P. Kianmajd, N. Franz, S. Bowers & B. Ludäscher. 2015. Provenance for explaining taxonomy alignments. In: Ludäscher, B. & B. Plale (Editors), Provenance and Annotation of Data and Processes. Revised Selected Papers of the 5th International Provenance and Annotation Workshop, IPAW 2014, Cologne, Germany, June 9-13, 2014. Lecture Notes in Computer Science 8628: 258-260. Available on-line here.
The first, fleshed out use case of the Euler/X project was published yesterday in PLoS ONE. This paper is a companion to the phylogenetic revision of the acalyptine weevil genus Perelleschus sec. Franz & Cardona-Duque (2013), and translates the 54 taxonomic concepts and 75 RCC-5 articulations provided in that paper into 13 logically consistent alignments and visualizations, with additional inferred articulations.
Franz, N.M., M. Chen, S. Yu, P. Kianmajd, S. Bowers & B. Ludäscher. 2015. Reasoning over taxonomic change: exploring alignments for the Perelleschus use case. PLoS ONE 10(2): e0118247. doi:10.1371/journal.pone.0118247. Available on-line here.
Very glad to see this one published; at the same time there are other use case papers in the pipeline (Andropogon, Primates). The particular motivation for this paper was to resolve sets of several small-scale yet taxonomically and phylogenetically complex input trees with the RCC-5 concept alignment approach and Euler/X toolkit. The paper is written in a “how to?” style, successively exploring and explaining the connections between the user-provided input constraints and the over-, under-, or well-specified reasoning outcomes. It deals with issues of logical consistency, input sufficiency, ambiguity, and alternative ways to align (parent) concepts in reference to either (1) their intensionally circumscribed properties (which may include synapomorphies) or (2) the ostensively indicated members. This corresponds to the program outlined in Franz & Thau (2010).
One reviewer wrote: “With an exceptionally suited use case, the complexity of taxonomic reasoning and its translation to machine processing are depicted in unprecedented form.” Our ultimate goal is to develop a widely applicable reference and linkage system for taxonomic products that human users create but which is actually optimized for computational processing – without compromising the Linnaean system whose services to humans are profoundly valuable.
Slides are up for the Euler/X Perelleschus (+) presentation at Entomology 2014.
Our lab has had an eventful joint ECN/ESA 2014 meeting and presentation/poster schedule of activities. Most presentations are now posted. Great meetings – even won recognition for our Twitter contributions.
Symposium – Harvesting the fruits of our labor: Utilizing collections databases to advance 21st century entomology.
1. Neil Cobb, Katja Seltmann & Nico Franz. 2014. The current state of arthropod biodiversity data: Addressing impacts of global change. Presentation.
“Part 1” – allowing for subsequent parts (e.g., for the blue/non-blue use case; now added at end of post). The relevant OpenTree background discussion is here:
There as several interesting and generally closely overlapping issues and views here. I will pick just (or mainly) the one introduced by Jonathan Rees (November 02, 2014).
“Here’s another example I’m struggling with: there are currently a couple of species in OTT that are misclassified as crustaceans instead of molluscs. When we fix this problem, there will be an incompatible ‘change’ in the membership of Arthropoda. Does this mean that the new group should get a new identifier? – after all its identity in some sense has changed. If so, annotations and OTU mappings linked to the old id have no home in the tree. It doesn’t get a new id with the current taxonomy generator, which assumes that names are tied uniquely to taxon concepts (with some exceptions), but with a more principled system where groups are defined by membership or phylogenetic hypotheses, it might. This would have an impact on OTU mappings and annotation carryover. I don’t have a good answer to this one, but am working on ways to anchor the semantics of ids.”
I am attempting to reproduce this under varying scenarios in the Euler/X toolkit. Starting simple, then expanding. I have created an initial scenario with two phylogenetic perspectives of the Ecdysozoa (molting animals) sec. ott1 (= OpenTree Topology at time =1; the later version) versus Ecdysozoa sec. ott0 (OTT at time = 0; the earlier version. Initially, an alignment with complete taxonomic congruence.