- I build smarter
software for science. Here's a list of my projects, my resumé,
my curriculum vitae, and some stuff I think is
Ontology for Major Histocompatibility RestrictionA paper describing our new ontology for MHC is now
histocompatibility complex is an important part of the
adaptive immune system, and MRO helps users search the IEDB for experiments involving different
classes of MHC molecules. We use ROBOT templates to build
MRO, and I think it's a good example of cutting edge ontology
development. You can browse the project source code at https://github.com/IEDB/MRO.
- The Command
LineOn Thursday I gave a presentation to the UofT Coders, a group of
graduate students in science at the University of Toronto who are
working together to improve their computer skills. It's part of
the Study Group initiative from the Mozilla Science Lab. I
talked about the Unix command line: what it does, how it works,
and why it's great.
- New OBO PURL
systemOn Monday I deployed
a new system for managing Persistent URLs (PURLs) for the OBO
community. It uses Git for configuration management and Apache
for serving the URLs. I'm pleased with the GitHub and Travis CI
integration for automatically testing all pull
- OBI Assay
TemplatesThe most recent
release of the Ontology for Biomedical Investigations uses
templates to define about half of the OBI assays. Template
spreadsheets make it much easier to edit large sets of terms
using the same pattern. The IEDB team put a lot of work into new
template files and we use ROBOT to translate the
templates into OWL format.
memory of Cesare RomagnoliLast week my
friend and colleague Cesare died. I owe my career to our
collaboration, and miss him dearly. I shared some of my memories
of him on the Rotman
- EDN-LD: A
simple linked data toolLast night I
presented a new project at the Clojure
Toronto Meetup. EDN-LD is designed to make working with
linked data simple, by using standard Clojure data structures and
the standard toolkit for working with those structures. Instead
of navigating a hierarchy of Java classes, you just use strings,
keywords, maps, and sets to work with your data, then convert the
results to a standard RDF format for other tools to use. I've
been using predecessors of EDN-LD in my work at IEDB for a few
years now, and now it's time to share.
- MHC Restriction
OntologyOn Friday the IEDB
team shared a preliminary version of a new ontology. Major
histocompatibility complex (MHC) is a family of protein
complexes that plays a crucial role in presenting molecules for
recognition by immune cells. Those molecules might be harmless
pieces of a healthy cell, or they could be fragments of a virus,
infectious bacterium, or toxin. MRO is not ready for wide use yet
(our term identifiers are UNSTABLE -- DON'T reuse them), but
we're already using it to drive our Allele
- ROBOT is an
OBO ToolFor the last few
months I've been working on a new project with Chris Mungall's
team at BerkeleyBOP.
ROBOT is a command-line tool for working with OBO ontologies,
designed to be a friendlier and more modular replacement for OWLTools. It's
open source, runs on Windows, Mac OS X, and Linux, and builds on
OWLAPI to do the
heavy lifting. ROBOT is still in early development, but it's
ready to try. We'd love to hear your thoughts.
Workshop SlideThe National Institutes of Health's Big Data to Knowledge (BD2K) program
has a number of parts. I was invited to their Workshop on
Community-Based Data and Metadata Standards Development this
week, and contributed to a panel on tools. Here's my slide.
We need better tools that specialize the linked data stack, and
don't require us to keep reinventing the wheel. But good tools
need good ergonomics and long-term maintenance, and that costs
Assay PatternsA quarter of the terms
in the Ontology for Biomedical
Investigations are assays, and they need to be cleaned up.
The IEDB team has been working on restructuring they assays we've
defined. I presented some of that work on the OBI call
gave a presentation on my explanation work for the Studies of
Translations, Ethics, and Medicine (STREAM) group at McGill
University in Montreal. We had a good discussion about extracting
structured knowledge from publications, and it was great to see
my friend Spencer
Ontology CoreOver the past month I've put in a lot of
hours into adding
documentation to the core terms in the OBO Relation
Ontology Now we have a new RO
- PSA 2014
PresentationToday I presented some
of my work on explanation at the Philosophy of Science Association
conference. The talk went pretty well, I think, and it was great
to see so many old friends.
- ontoden-deployI spent a couple of days this week using Ansible and Vagrant to get Ontobee and OntoFox
running in a virtual machine. The code is up on
created a Google Doc to track the discussions in our workshop at
2014 in Houston this morning.
Tutorial v0.1There's a lot to
cover, and it's not done yet, but I've made substantial additions
to the OBO tutorial, and I'm calling it "v0.1".
- OBI Term
GuidelinesI wrote some new
guidelines for submitting terms to OBI. Most of the advice
applies to any OBO ontology.
TutorialI've started writing
a tutorial for scientists working with OBO ontologies. It's about
half done now, and we'll be using some of this material in the OBO
Tutorial at ICBO 2014.
- OBI Release
2014-08-18This OBI release
includes three new terms.
Peirce and the Philosophy of Medical ImagingThis morning
Cesare, Nick, and I presented on Peirce and radiology at the The Charles S.
Peirce International Centennial Congress. The breadth and
scope of the presentations here is impressive!
and Imaging Informatics WorkshopHeiner Oberkampf and I just presented his recent
work applying OBI to radiology reports. You can download
our slides and see the full
Tutorial and Workshop at ICBO 2014I'm helping to organize a Tutorial
on Open Biomedical Ontologies at the 5th International Conerence on Biomedical
- Rotman GRA
and I are grateful to the Rotman Institute for funding a graduate
research assistant position for our project "The Ontology and
Epistemology of Medical Imaging". We're even happier that Nick
Nash has accepted our invitation to join the project!
Classes-Only VersionI helped
arrange today's release of BFO, with a new classes-only
perils of managing OWL in a version control systemChris Mungall has sparked a good discussion about
how OWL/XML doesn't play nicely with version control
- OBI Release
2014-03-29A new version of
OBI has been released -- not an April Fool's joke!
HackathonToday I worked on deployment
automation for the Ontohub
project as part of the OntoLog hackathon.
- OBI Release
ProcessOBI has adopted a
new process for more frequent releases. I'm the first OBI
- OBI Release
2013-10-25This is the first
release of OBI to use the new build tool, and the first to
Foundry Technical Working GroupThe Working Group coordinates on the infrastructure
and technical policies of the OBO Consortium. Today I'm its
Build ToolI've written a new
build tool for testing and releasing OBI.
- BFO 1.2
ProposalI think that the Basic Formal Ontology needs to
release a version 1.2 ...
- DissertationOne month ago I successfully defended my doctoral
- SubmissionToday I submitted my doctoral
- Philosophy of
ScienceWhat is philosophy
of science, exactly?
- About this
SiteIt's been a long
time coming, but I've finally updated my homepage...