From: MARIA H. Z. <asm...@gm...> - 2014-06-25 21:13:15
|
Hi Chris. Thanks a lot for your comments. The aim of this project is to represent all information related with DDIs in a single ontology. To do this we have reused information from different resources and integrated them in a common conceptual model. Please, find below my comments, Best regards, Maria > > > Hi Maria, > > It looks like DINTO is an automatic translation of DrugBank - is this > correct? > DrugBank is one of our main information resources. Therefore, some information in DINTO has as original source DrugBank. This includes DDIs and their mechanisms and drug-protein relationships. > It's great that you are bringing in classes from CHEBI, but it looks like > many things in the DINTO ID space belong elsewhere. I don't think anyone > needs yet another set of IDs for genes and proteins. > Drugs were originally imported from the ChEBI ontology – specifically, we selected those chemical entities described to have a role ‘drug’ – to then represent DDIs between them. However, some drugs in DrugBank were not in our ChEBI dataset and were included in the ontology as new chemical entity classes. We reused other ontologies covering information regarding with the DDI domain, as well. Pharmacokinetic parameters come from the Pharmacokinetics ontology (PKO) and information resources come from the Biomedical Resource Ontology (BRO). Regarding with proteins, they were included to represent drug-protein relationships that will allow the inference of new DDIs in basis of their mechanisms in future inferred versions of the ontology. I think that you are right and that it would not be necessary to provide new IDs for them. We maintain the UniProtID so we can modify this in future versions while maintaining those drug-protein relationships already represented in the ontology. > > How do you see the long term evolution of DINTO? Specifically with respect > to which portions of the ontology you see as unique contributions to OBO, > which parts are imported, and which parts are edges brought in by automatic > translation from external databases? > The current version of DINTO has been designed to continue growing and importing DDI related information. Future versions of the ontology will integrate DDI information from other resources (both databases and ontologies). The next steps will be to integrate current DDIs represented in DINTO with those described in KEGG database and the NDFRT. Future work will include representation of other relevant aspects of DDIs, such as drug effects – both therapeutic and adverse effects – in the ontology. To do this, one adverse drug reactions ontology (e.g. AERO or OAE) will be reused and drug-ADR relationships will be obtained from the database SIDER. > > On Tue, Jun 24, 2014 at 6:01 AM, MARIA HERRERO ZAZO <mh...@pa...> > wrote: > > > Dear OBO Foundry members, > > > > > > > > At the *Advanced Databases Group* (LaBDa) (*University Carlos III of > > Madrid*) we have been working on a new biomedical ontology: the > *Drug-drug > > Interaction Ontology* *(DINTO)*. > > > > > > > > Drug-drug interactions (DDIs) form a significant risk group for adverse > > effects associated with pharmaceutical treatment. These interactions are > > often reported in the literature, however, they are sparsely represented > in > > machine-readable resources, such as online databases, thesauri or > > ontologies. While ontologies for drugs have proliferated in recent years, > > there is no ontology capable of describing and categorizing drug-drug > > interactions. Moreover, there is no artifact that represents all the > > possible mechanisms that can lead to a DDI. To fill this gap we have > > created DINTO, an ontology for drug-drug interactions and their > mechanisms. > > > > > > > > In the development of this project we are following best practices laid > > out by the OBO Foundry. This project has been developed in collaboration > > with the ChEBI team at the European Bioinformatics Institute (EBI) and > the > > ontology is free for use and available at > https://code.google.com/p/dinto/. > > Moreover, it is aligned with the OBO Foundry ontology Basic Formal > Ontology > > (BFO). Therefore, we would most appreciate your kind consideration in > > registering DINTO as an OBO Foundry member ontology. > > > > > > > > As a first step to join the OBO Foundry community we would like to > reserve > > the name space for ?The Drug-drug Interaction Ontology" with abbreviation > > of "DINTO". > > > > > > > > Required information is described below: > > > > > > > > - requested prefix: DINTO > > > > - ontology title: The Drug-drug Interaction Ontology > > > > - download link: > https://googledrive.com/host/0B-7Po9tR1KLUNkpNdmFEcG44RjA > > > > - ontology home page: https://code.google.com/p/dinto/ > > > > - person of contact: Maria Herrero > > > > - person of contact email: mh...@pa... > > > > - ontology discussion mailing list: > > https://groups.google.com/forum/#!forum/dinto-discussion > > > > - ontology issue tracker link: > https://code.google.com/p/dinto/issues/list > > > > - PURL user ID: herrerozazo > > > > > > > > First steps in the creation of DINTO are described in Herrero-Zazo, M., > > Hastings, J., Segura-Bedmar, I., Croset, S., Martinez, P., & Steinbeck, > C. > > (2013). *An ontology for drug-drug interactions*. In *6th International > > Workshop on Semantic Web Applications and Tools for Life Sciences > (SWAT4LS)*. > > Edinburgh, UK. > > > > > > > > Kind regards, > > > > > > > > Maria Herrero Zazo > > > > Laboratorio de Bases de Datos Avanzadas (LaBDa) > > > > Universidad Carlos III de Madrid > > > > http://labda.inf.uc3m.es/doku.php?id=en:inicio > > > > > > > ------------------------------------------------------------------------------ > > Open source business process management suite built on Java and Eclipse > > Turn processes into business applications with Bonita BPM Community > Edition > > Quickly connect people, data, and systems into organized workflows > > Winner of BOSSIE, CODIE, OW2 and Gartner awards > > http://p.sf.net/sfu/Bonitasoft > > _______________________________________________ > > Obo-discuss mailing list > > Obo...@li... > > https://lists.sourceforge.net/lists/listinfo/obo-discuss > > > > > -------------- next part -------------- > An HTML attachment was scrubbed... > > |