HCLSIG/PharmaOntology/Meetings/2009-08-27 Conference call
Conference Details
- Date of Call: Thursday August 27 2009
- Time of Call: 9:00am - 1pm ET
- Dial-In #: +1.617.761.6200 (Cambridge, MA)
- Dial-In #: +33.4.89.06.34.99 (Nice, France)
- Dial-In #: +44.117.370.6152 (Bristol, UK)
- Participant Access Code: 42572 ("HCLS2").
- IRC Channel: irc.w3.org port 6665 channel #HCLS2 (see W3C IRC page for details, or see Web IRC)
- Mibbit instructions: go to http://www.mibbit.com/chat and click the server link. Enter irc.w3.org:6665 into that box, enter a nickname, and enter #HCLS2 for the channel
- Duration: 4h
- Convener: Susie
Agenda
- Review of Classes and Class Definitions
Minutes
Translational Medicine Ontology
August 27, 2009
Attendees: Susie, Colin, Christi, Elgar, Chris, Michel, Anja, Julia, Trish
Efficacy
Colin: too low level
Colin: this will probably do. It’s a collection of outcomes
Susie: this has an element of how active the drug is… relationships between efficacy and outcome
Colin: depends on patient
Decision: keep Efficacy
Excipient
Susie: this goes back to discussion on terms to describe drug (inactive, active, etc.)
Elgar: compound, drug, pharma product, brand were term from last time. In discussion from drug, have not concluded that drug=active ingredient
Elgar: do we need inactive/active? Yes. Need drug or active ingredient. Need to know which are synonyms.
Elgar: this and active ingredient seem to be synonyms
Colin: agree that we need inactive ingredient
Susie: it does need to be captured
Experiment
Colin: experiment is collection of assays
Susie: would be good to hear from Bosse
Elgar: considering study, experiment, investigation... how relate
Colin: study and investigation the same?
Colin: experiment may be collection of assays?
Susie: want to expose the conclusions
Colin: defer to OBI?
Michel: OBI has something like assay, but not clinical trial. could potentially be added.
ElgarPichler concepts in this space: study investigation experiment assays test trial
Elgar: investigation as main term
Michel: OBI has investigation w/ hypothesis generating investigation and hypoth driven investigation
Michel: should be able to add what we need to OBI
Experimental Output
Colin: same as experiment
Experimental Result
Colin: same as experiment
Fragment
Colin: should drop, too specific
Susie: might want something to reflect that it's part of a chemical series
Colin: punt to Chebi
Elgar: substructure search for related compounds
Elgar: would be use for this
Gene
See spreadsheet
Gene Expression
Colin: why have this here?
Elgar: Gene Expression Profile collected by Elgar; all suggestions
Elgar: important is that certain phenotype has genetic profile
Elgar: these are common ways of classifying patient
Colin: phenotypes that don’t arise from gene expression?
Susie: yes
Colin: is tennis elbow phenotypic?
Susie: phenotype is what it looks like, rather than why it got there
Susie: likes gene expression for classifying patients
Colin: genotypic handled by things like SNP
Susie: gene expression on it’s own isn’t quite what we want
Elgar: do we need phenotype and genotype?
Elgar: if we need phenotype and genotype, then we should go down to what type and what things we are actually looking at for phenotype and genotype
Elgar: revisit how gene expression fits in?
Gene Expression Phenotype
Colin: prefer phenotype
Gene Expression Profile
Gene Expression Profiling
Genomic Sequence
Colin: pull from SO
Immune Response
Susie: there are many ways a human would respond to a drug
Colin: there are many responses. This is too specific.
Colin: patient response?
Susie: don’t think we captured drug metabolism?
Colin: yes , in 38
Colin: add response instead of immune response
Susie: product response?
Susie: similar to outcome, but could include toxicology
Elgar: could be effect on cell line
Christi: subject could include that
Inactive Ingredient
Michel: prefer excipient
Michel: not aware of term for active ingredient
Information
Colin: vague
Intervention
Colin: rename to perturbation?
Elgar: has positive intent, perturbation is just what’s going in/out and what’s measurable
Elgar: intervention is better known in the field
Colin: keep intervention
Market
Susie: important, but would choose market opportunity instead
Colin: market opportunity is a gap in the market
Susie: would unmet need be more appropriate?
Market Opportunity
Colin: may be same as market
Material Entity
Colin: in BFO, we don’t need
Measure
Susie: loose
Colin: vague
Metabolic Pathway
Michel: suggests some pathway
Michel: pathway is important to explain the mechanism, how something occurs
Susie: replace with biological pathway?
Colin: use pathway
Metabolism
Susie: drug metabolism is the focus
Michel: and how the drug affects a normal metabolism
Susie: drug effect covered by response?
Elgar: drug metabolizing enzyme would be covered by drug metabolism
Elgar: need metabolite to further specify phenotype?
Colin: yes
Molecular Structure
Susie: need something about structure to have concept of chemical families
Colin: Chebi doesn’t deal with non-existent molecules
Colin: say 3D molecular structure instead
Observation
Susie: connects into C-DISC, may be in HL7
Organ
Elgar: refer to FMA
Outcome
Susie: patient outcome would be preferred, depends on whether we want to cover experiments too
Susie: patient outcome is a common phrase in pharma
Elgar: finding observation outcome results…. Same category
Elgar: concept group: (subject, patient) outcome, observation, diagnosis, ...
Elgar: diagnosis too when look at outcome?
Colin: complicated, need to think about more
Susie: reference health outcome
Patent
Susie: market opportunity depends on existing patents
Elgar: agree
Pathway
Patient
Group: Prefer subject, if needed, can sub-classify
Michel: this is clinical setting
Patient Diagnosis
Elgar: not sure why need the ‘patient’ on there
Patient Prognosis
Payer
Susie: need to include the organization that pays for the drug
Christi: includes patient
Elgar: how far down does the ontology stretch?
Elgar: need to look at prognosis too
Person
Pharmaceutical Brand
Pharmaceutical Product
Pharmacological Activity
Elgar: describe by active/inactive and interactions between protein and compound
Christi: like the mechanism of action? (the how)
Michel: this is functional-type
Pharmacophore
Michel: important with respect to drug design and development, a mathematical model
Susie: seems detailed for this ontology
Michel: important for the drug description
Phenotype
Polypharmacology
Elgar: could interact with different sites, true for most of the compounds
Elgar: too specific?
Michel: don’t need a set concept, can refer to activity
Population
Susie: how tie into epidemiology and longitudinal studies? Do we have a term to represent this set of people?
Elgar: collection of subjects in the study?
Elgar: get to study population by grouping on various qualities?
Susie: study population is good
Project
Protein
Protein Sequence
Elgar: make attribute of protein, gene
Receptor
Susie: we would say it binds to protein, but not specific receptor
Michel: may need for signaling pathways in drug targeting
Referral
Michel: too much in the health care domain
Regulator
Michel: Chebi has ‘effector’ for the biological context
Christi: prefer to call the approval body for drugs a Regulatory Authority
Regulatory Authority (new)
Repressor
Elgar: maybe Chebi (activator or inhibitor)
Resource
Risk
Colin: needs to be taken into account or something that has happened before
Susie: risk of developing a drug
Colin: subject risk and financial risk
Elgar: don’t need core financial risk
Colin: this will be subject risk
Role
Colin: pull from elsewhere?
Christi: not as valuable in a generic sense
Sample
Michel: OBI has a term ‘sample population’
Scaffold
Colin: too specific
Sequence
Michel: use the sequence ontology (SO)
Side Effect
Susie: Need either side effect or adverse event
Michel: not all side effects are adverse events
All: keep
Signaling
Colin: Too specific
Signaling Pathway
Michel: use pathway
SNP
Colin: adding copy number variation
Solution
All: delete
Symptom
Susie: need sign as well
Elgar: need sign, symptom, observation defined
Chris: from clinical – symptoms are patient, signs are visible finding on clinical examination
Observation -> recording signs
Synthesis
Colin: too domain specific
Target
Test
Elgar: replace
Susie: study, assay, or investigation preferred
Therapy
Chris: Treatment and therapy may be used somewhat interchangeably
Elgar: like Bosse’s comment about using intervention
Therapy Cost Benefit
Tissue
Susie: similar to organ
Michel: organs in FMA
Toxicity
Michel: quantifiable measures or likelihood
Michel: predictive outcome
Toxicology
Colin: out (field of study)
Treatment
Treatment Management Plan
Michel: a specification to undertake treatment
Colin: we have clinical protocol
Michel: need to understand what the clinical protocol is meant to specify
Colin: merge into clinical protocol
Unit
Clinical Protocol (new)
Safety
Evidence
Susie: ties into tailored therapeutics
Chris: UK evidence database, quantifiable
Pharmacogenetics
Susie: still need to capture concepts around tailoring of drugs
Lead
Colin: role played by a molecule, defer to Cheminformatics ontology
Pathology
Susie: covered by disease
PK/PD
Colin: sounds ‘field-y’
Chris: refers to metabolism or drug interaction
Susie: may still want to consider pharmacodynamics
Syndrome
Elgar: collection of signs, synonymous to disease?
Cost
Susie: any ontologies with this?
Trish found a few
Physiology
Molecule
Mechanism of Action
Heredity
Chris: interested in the generations and how they metabolize drug
Michel: not sure these are necessary for what we are trying to do
Event
Finding
Susie: vague
Michel: part of scientific discourse, outcomes of experiments can be used for evidence
Trish: in OBI
Comparative Effectiveness
Chris: initially placebo, then to other drugs in the class
Susie: keep and look at in the context of outcomes
Sign (new)
Stratification
Susie: patient grouping
Prognosis
Inclusion/Exclusion Criterion
Christi: defer to trial ontology?
Arm
- Changed to arm of clinical study
- Defer to trial ontology
Experiment Design
Image
Christi: seems like evidence, e.g. image of slide of cells
Susie: not the final term we would want to include
Colin: could be evidence
Class (of compound)
Colin: defer to Chebi
ADME
Susie: covered by drug metabolism
Chris: that would give us better control over the information, classify by transdermal, sublingual, etc.
Colin: defer to domain ontology
Elgar: this is like PK/PD – a bundling
Baseline
Locus
Polymorphism
Genotype
Gene List/ Gene Signature
Susie: for gene expression analysis
Susie: don’t think we need it
Validation
Susie: don’t think we need
Evidence
Hypothesis
Medical History
Cognition
Susie: may be able to cover with disease
- Replaced with Biomedical Measure
Geography
- Replaced by country
Environment
Elgar: includes geographical factors
Susie: include lifestyle factors?
Trish: keep separate
Chris: may include upbringing
Elgar:
Subject defined by phenotype
Environment influences phenotype
Lifestyle
Susie: important
Sample Restriction
Unclear
Statistical Power
Chris: measure of credibility in claiming you have accepted your HO or HA, your hypothesis; power in the study
Colin: covered by evidence?
Response by Subject to Drug
Metabolite
Generic
Susie: for comparative analysis and cost/benefit analysis
Susie: generic drug name should be covered
Trish: a synonym
Copy Number Variation (new)
Next steps:
Look at clean version of spreadsheet, categorize by keep/remove Add definitions for new terms
Action Items:
Elgar will restructure spreadsheet and do conceptual grouping Susie will go into BioPortal and look for definitions