microarray-ontol-digest Friday, November 16 2001 Volume 01 : Number 014 ---------------------------------------------------------------------- Date: Fri, 02 Nov 2001 18:12:05 +0100 From: Martin Hofmann Subject: Re: [microarray-ontol] feedback on ontology development Dear Chris, concerning clinical history: how does ICD-9 and SNOMED deal with this? I mean: can you take over categories from these systems? Or just point towards them? I find the concept of "pointing" into other knowledge domains (other ontologies) rather attractive - simply because we cannot cover all the different fields. As far as compounds are concerned, pointing into the Merck Index is a first step. I am not sure at the moment, but if the Merck index contains chemical abstract indices this would be already a big stept forward. However, there are real professional compound registration systems out there which - unfortunately - are commercial but they are also the best systems available to describe, to search and to register compounds. Maybe I can give a few examples at MGED 4 about how the world of chemistry could be linked into the microarray ontology .... So far my comments. Is there anybody out there who has other ideas on the representation of the chemistry world? best wishes, Martin Chris Stoeckert wrote: > Dear Group, > Some things I've been thinking about regarding the MGED Ontology that I > would appreciate your thoughts on. First let me point you at the html > page: > http://www.cbil.upenn.edu/Ontology/biomaterial11.html > > ClinicalHistory subclasses: Right now the only attributes for > ClinicTreatmentHistory, CurrentDiseaseHistory, and PastMedicalHistory > are those inherited from ClinialHistory (i.e., lab values and clinical > information). This doesn't strike me as sufficient but I'm not sure what > else to add. Any suggestions? > > Compund: Compound is "A drug, solvent, chemical, etc., that can be > measured. " I'm changing this class to be a subclass of OntologyEntry so > that people can point to the Merck Index or catalogs. Any objections? > > OntologyEntry: Not all ontologies are on-line and can be specified as a > DatabaseEntry. I'm thinking of adding attributes: has_citation, > resource_name. > OK with you? > > BehavioralStimulus: Currently this has no attributes except has_protocol > inherited from Treatment. I'm thinking of expanding the attributes to > include > stimulus_type. Instances (individuals) for stimulus_type would be > avoidance, reward, deprivation (others?). stimulus_name could also be an > attribute but I' m not sure how to describe instances like "shock" or > "sleep" and also use a class like Compound. thoughts? > > Cheers, > Chris > > Chris Stoeckert, Ph.D. > Research Associate Professor, Dept. of Genetics > Center for Bioinformatics, University of Pennsylvania > 418 Guardian Dr., Philadelphia, PA 19104 > Ph:215-573-4409 FAX:215-573-3111 - -- Dr. Martin Hofmann Executive Project Director LION bioscience AG Im Neuenheimer Feld 515 69120 Heidelberg Germany phone: (+49)-6221-4038-120 cell.: (+49)-(0)173-9303828 fax: (+49)-6221-4038-401 e-mail: martin.hofmann@lionbioscience.com www: http://www.lionbioscience.com ------------------------------ Date: Mon, 05 Nov 2001 10:44:38 +0000 From: susanna sansone Subject: Re: [microarray-ontol] feedback on ontology development Dear all, Re: the chemistry world. IUPAC has been involved in producing naming systems and terminology recommendations for a very long time. Many IUPAC and IUBMB recommendations are available via Dr Gerry Moss' website at Queen Mary College (www.chem.qmul.ac.uk/iupac), and a more complete (PDF) list is at http://www.rsc.org/pdf/general/append.pdf IUPAC is also hoping to initiate a specific XML project via the proposed meeting in Cambridge, UK, at the end of next January. IUPAC has 3 ongoing projects (http://iupac.chemsoc.org/news/archives/2000/NRT_Report.html): * The Chemical Identifier project (IChI). The aim of the project is to establish a unique label, the IUPAC Chemical Identifier (IChI), which would be a non-proprietary identifier for chemical substances. This is now in progress (see http://www.iupac.org/projects/2000/2000-025-1-050.html ) but has not yet produced a beta-version. They expect this to appear before the end of the year. * Chemical Markup Language (CML) For IUPAC, work in this area is coordinated by the Committee on Printed and Electronic Publications. Prof.Peter Murray-Rust was largely responsible for the early development of CML (http://www.oasis-open.org/cover/cml.html) and he is still actively involved with it. We are in contact with him and he will keep us updated. Hope is useful. Regards, Susanna Martin Hofmann wrote: > > Dear Chris, > concerning clinical history: how does ICD-9 and SNOMED deal with this? I > mean: can you take over categories from these systems? Or just point towards > them? I find the concept of "pointing" into other knowledge domains (other > ontologies) rather attractive - simply because we cannot cover all the > different fields. > As far as compounds are concerned, pointing into the Merck Index is a first > step. I am not sure at the moment, but if the Merck index contains chemical > abstract indices this would be already a big stept forward. However, there > are real professional compound registration systems out there which - > unfortunately - are commercial but they are also the best systems available > to describe, to search and to register compounds. Maybe I can give a few > examples at MGED 4 about how the world of chemistry could be linked into the > microarray ontology .... > > So far my comments. Is there anybody out there who has other ideas on the > representation of the chemistry world? > > best wishes, > > Martin > > Chris Stoeckert wrote: > > > Dear Group, > > Some things I've been thinking about regarding the MGED Ontology that I > > would appreciate your thoughts on. First let me point you at the html > > page: > > http://www.cbil.upenn.edu/Ontology/biomaterial11.html > > > > ClinicalHistory subclasses: Right now the only attributes for > > ClinicTreatmentHistory, CurrentDiseaseHistory, and PastMedicalHistory > > are those inherited from ClinialHistory (i.e., lab values and clinical > > information). This doesn't strike me as sufficient but I'm not sure what > > else to add. Any suggestions? > > > > Compund: Compound is "A drug, solvent, chemical, etc., that can be > > measured. " I'm changing this class to be a subclass of OntologyEntry so > > that people can point to the Merck Index or catalogs. Any objections? > > > > OntologyEntry: Not all ontologies are on-line and can be specified as a > > DatabaseEntry. I'm thinking of adding attributes: has_citation, > > resource_name. > > OK with you? > > > > BehavioralStimulus: Currently this has no attributes except has_protocol > > inherited from Treatment. I'm thinking of expanding the attributes to > > include > > stimulus_type. Instances (individuals) for stimulus_type would be > > avoidance, reward, deprivation (others?). stimulus_name could also be an > > attribute but I' m not sure how to describe instances like "shock" or > > "sleep" and also use a class like Compound. thoughts? > > > > Cheers, > > Chris > > > > Chris Stoeckert, Ph.D. > > Research Associate Professor, Dept. of Genetics > > Center for Bioinformatics, University of Pennsylvania > > 418 Guardian Dr., Philadelphia, PA 19104 > > Ph:215-573-4409 FAX:215-573-3111 > > -- > Dr. Martin Hofmann > Executive Project Director > LION bioscience AG > Im Neuenheimer Feld 515 > 69120 Heidelberg > Germany > phone: (+49)-6221-4038-120 > cell.: (+49)-(0)173-9303828 > fax: (+49)-6221-4038-401 > e-mail: martin.hofmann@lionbioscience.com > www: http://www.lionbioscience.com - -- ******** Assunta Susanna Sansone, PhD Microarray Informatics Team EBI - The European Bioinformatics Institute EMBL Outstation - Hinxton Wellcome Trust Genome Campus Cambridge CB10 1SD, UK email: sansone@ebi.ac.uk direct: +44 (0)1223 494 691 fax: +44 (0)1223 494 468 http://www.ebi.ac.uk/microarray ******** ------------------------------ Date: Wed, 7 Nov 2001 11:23:33 -0500 From: Chris Stoeckert Subject: Re: [microarray-ontol] feedback on ontology development Hi Martin, > concerning clinical history: how does ICD-9 and SNOMED deal with this? I > mean: can you take over categories from these systems? Or just point > towards > them? I find the concept of "pointing" into other knowledge domains > (other > ontologies) rather attractive - simply because we cannot cover all the > different fields. > ClinicalHistory does point to external resources through the attribute has_clinical_information(ClinicalInformation) where ClinicalInformation is a BiosourceOntologyEntry. I think ICD-9 and SNOMED (perhaps GALEN as well) could be made instances (individuals in OILed) of ClinicalInformation. ICD-9 is used to code and classify mortality data from death certificates and so may not be appropriate for ClinicTreatmentHistory, PastMedicalHistory, and CurrentDiseaseHistory. SNOMED RT lists the following root hierarchies on it's web site: * Findings, Conclusions and/or Assessments * Procedures * Body Structures * Biological Functions * Living Organisms * Substances (Chemicals & Drugs) * Physical Agents, Activities and/or Forces * Occupations * Social Context * Modifiers/Linkage Terms and/or Qualifiers which certainly would be appropriate for these subclasses of ClinicalHistory (but SNOMED requires licensing). GALEN is open and "The GALEN CORE Model for representation of the Common Reference Model for Procedures contains the building blocks for defining procedures – the anatomy, surgical deeds, diseases, and their modifiers used in the definitions of surgical procedures. " Are there others that people can recommend? My question however was whether ClinicTreatmentHistory, PastMedicalHistory, and CurrentDiseaseHistory required any additional attributes. For example, should I add attributes: "where_treated(Contact)", "who_treated(Contact)", and "date" to indicate where the treatment or clinical test was done, who saw the patient, and when. Cheers, Chris ------------------------------ Date: Wed, 7 Nov 2001 12:07:11 -0500 From: Chris Stoeckert Subject: Re: [microarray-ontol] feedback on ontology development Susanna, Absolutely. "The purpose of the LOINC database is to facilitate the exchange and pooling of results, such as blood hemoglobin, serum potassium, or vital signs, for clinical care, outcomes management, and research." so this fits nicely. Michael Ashburner pointed this out to me awhile back and I never got around to including it on my list of resources. Thanks! Chris On Wednesday, November 7, 2001, at 11:58 AM, susanna sansone wrote: > Could LOINC database be useful for the scope? > LOINC is HL7’s preferred code system and it is free > http://www.regenstrief.org/loinc/ > > Regards, > Susanna > > Chris Stoeckert wrote: >> >> Hi Martin, >>> concerning clinical history: how does ICD-9 and SNOMED deal with >>> this? I >>> mean: can you take over categories from these systems? Or just point >>> towards >>> them? I find the concept of "pointing" into other knowledge domains >>> (other >>> ontologies) rather attractive - simply because we cannot cover all the >>> different fields. >>> >> ClinicalHistory does point to external resources through the attribute >> has_clinical_information(ClinicalInformation) where ClinicalInformation >> is a BiosourceOntologyEntry. I think ICD-9 and SNOMED (perhaps GALEN as >> well) could be made instances (individuals in OILed) of >> ClinicalInformation. ICD-9 is used to code and classify mortality data >> from death certificates and so may not be appropriate for >> ClinicTreatmentHistory, PastMedicalHistory, and CurrentDiseaseHistory. >> SNOMED RT lists the following root hierarchies on it's web site: >> * Findings, Conclusions and/or Assessments >> * Procedures >> * Body Structures >> * Biological Functions >> * Living Organisms >> * Substances (Chemicals & Drugs) >> * Physical Agents, Activities and/or Forces >> * Occupations >> * Social Context >> * Modifiers/Linkage Terms and/or Qualifiers >> which certainly would be appropriate for these subclasses of >> ClinicalHistory (but SNOMED requires licensing). GALEN is open and "The >> GALEN CORE Model for representation of the Common Reference Model for >> Procedures contains the building blocks for defining procedures – the >> anatomy, surgical deeds, diseases, and their modifiers used in the >> definitions of surgical procedures. " >> >> Are there others that people can recommend? >> >> My question however was whether ClinicTreatmentHistory, >> PastMedicalHistory, and CurrentDiseaseHistory required any additional >> attributes. For example, should I add attributes: >> "where_treated(Contact)", "who_treated(Contact)", and "date" to >> indicate where the treatment or clinical test was done, who saw the >> patient, and when. >> >> Cheers, >> Chris > > -- > ******** > Assunta Susanna Sansone, PhD > > Microarray Informatics Team > > EBI - The European > Bioinformatics Institute > EMBL Outstation - Hinxton > Wellcome Trust Genome Campus > Cambridge CB10 1SD, UK > > email: sansone@ebi.ac.uk > direct: +44 (0)1223 494 691 > fax: +44 (0)1223 494 468 > http://www.ebi.ac.uk/microarray > ******** > ------------------------------ Date: Fri, 16 Nov 2001 16:28:13 +0000 From: susanna sansone Subject: Re: [microarray-ontol] feedback on ontology development Chris, Re: the chemistry world by the way I forgot to mention ChemIDplus search system in my last email. It is free at http://chem.sis.nlm.nih.gov/chemidplus/ It provides access to structure and nomenclature authority files used for the identification of chemical substances cited in National Library of Medicine (NLM) databases. The database contains over 350,000 chemical records, direct links to many biomedical resources at NLM and on the Internet for chemicals of interest+ Name and Synonyms, CAS Registry Number, ID, Molecular Formula, Classification Code, Locator Code, and Structure (an information icon lists the locator sources and data element). Have you had a look at it? Regards, Susanna susanna sansone wrote: > > Dear all, > > Re: the chemistry world. > > IUPAC has been involved in producing naming systems and terminology > recommendations > for a very long time. Many IUPAC and IUBMB recommendations are available > via > Dr Gerry Moss' website at Queen Mary College > (www.chem.qmul.ac.uk/iupac), > and a more complete (PDF) list is at > http://www.rsc.org/pdf/general/append.pdf > IUPAC is also hoping to initiate a specific XML project via the proposed > meeting in > Cambridge, UK, at the end of next January. > > IUPAC has 3 ongoing projects > (http://iupac.chemsoc.org/news/archives/2000/NRT_Report.html): > > * The Chemical Identifier project (IChI). > > The aim of the project is to establish a unique label, the IUPAC > Chemical Identifier (IChI), which would be a non-proprietary identifier > for chemical substances. > This is now in progress (see > http://www.iupac.org/projects/2000/2000-025-1-050.html ) > but has not yet produced a beta-version. They expect this to appear > before the > end of the year. > > * Chemical Markup Language (CML) > > For IUPAC, work in this area is coordinated by the Committee on > Printed and Electronic Publications. > Prof.Peter Murray-Rust was largely responsible for the early development > of CML (http://www.oasis-open.org/cover/cml.html) > and he is still actively involved with it. > We are in contact with him and he will keep us updated. > > Hope is useful. > > Regards, > > Susanna > > Martin Hofmann wrote: > > > > Dear Chris, > > concerning clinical history: how does ICD-9 and SNOMED deal with this? I > > mean: can you take over categories from these systems? Or just point towards > > them? I find the concept of "pointing" into other knowledge domains (other > > ontologies) rather attractive - simply because we cannot cover all the > > different fields. > > As far as compounds are concerned, pointing into the Merck Index is a first > > step. I am not sure at the moment, but if the Merck index contains chemical > > abstract indices this would be already a big stept forward. However, there > > are real professional compound registration systems out there which - > > unfortunately - are commercial but they are also the best systems available > > to describe, to search and to register compounds. Maybe I can give a few > > examples at MGED 4 about how the world of chemistry could be linked into the > > microarray ontology .... > > > > So far my comments. Is there anybody out there who has other ideas on the > > representation of the chemistry world? > > > > best wishes, > > > > Martin > > > > Chris Stoeckert wrote: > > > > > Dear Group, > > > Some things I've been thinking about regarding the MGED Ontology that I > > > would appreciate your thoughts on. First let me point you at the html > > > page: > > > http://www.cbil.upenn.edu/Ontology/biomaterial11.html > > > > > > ClinicalHistory subclasses: Right now the only attributes for > > > ClinicTreatmentHistory, CurrentDiseaseHistory, and PastMedicalHistory > > > are those inherited from ClinialHistory (i.e., lab values and clinical > > > information). This doesn't strike me as sufficient but I'm not sure what > > > else to add. Any suggestions? > > > > > > Compund: Compound is "A drug, solvent, chemical, etc., that can be > > > measured. " I'm changing this class to be a subclass of OntologyEntry so > > > that people can point to the Merck Index or catalogs. Any objections? > > > > > > OntologyEntry: Not all ontologies are on-line and can be specified as a > > > DatabaseEntry. I'm thinking of adding attributes: has_citation, > > > resource_name. > > > OK with you? > > > > > > BehavioralStimulus: Currently this has no attributes except has_protocol > > > inherited from Treatment. I'm thinking of expanding the attributes to > > > include > > > stimulus_type. Instances (individuals) for stimulus_type would be > > > avoidance, reward, deprivation (others?). stimulus_name could also be an > > > attribute but I' m not sure how to describe instances like "shock" or > > > "sleep" and also use a class like Compound. thoughts? > > > > > > Cheers, > > > Chris > > > > > > Chris Stoeckert, Ph.D. > > > Research Associate Professor, Dept. of Genetics > > > Center for Bioinformatics, University of Pennsylvania > > > 418 Guardian Dr., Philadelphia, PA 19104 > > > Ph:215-573-4409 FAX:215-573-3111 > > > > -- > > Dr. Martin Hofmann > > Executive Project Director > > LION bioscience AG > > Im Neuenheimer Feld 515 > > 69120 Heidelberg > > Germany > > phone: (+49)-6221-4038-120 > > cell.: (+49)-(0)173-9303828 > > fax: (+49)-6221-4038-401 > > e-mail: martin.hofmann@lionbioscience.com > > www: http://www.lionbioscience.com > > -- > ******** > Assunta Susanna Sansone, PhD > > Microarray Informatics Team > > EBI - The European > Bioinformatics Institute > EMBL Outstation - Hinxton > Wellcome Trust Genome Campus > Cambridge CB10 1SD, UK > > email: sansone@ebi.ac.uk > direct: +44 (0)1223 494 691 > fax: +44 (0)1223 494 468 > http://www.ebi.ac.uk/microarray > ******** - -- ******** Assunta Susanna Sansone, PhD Microarray Informatics Team EBI - The European Bioinformatics Institute EMBL Outstation - Hinxton Wellcome Trust Genome Campus Cambridge CB10 1SD, UK email: sansone@ebi.ac.uk direct: +44 (0)1223 494 691 fax: +44 (0)1223 494 468 http://www.ebi.ac.uk/microarray ******** ------------------------------ End of microarray-ontol-digest V1 #14 *************************************