search

sorgerlab / indra

INDRA (Integrated Network and Dynamical Reasoning Assembler) is an automated model assembly system interfacing with NLP systems and databases to collect knowledge, and through a process of assembly, produce causal graphs and dynamical models.
http://indra.bio
BSD 2-Clause "Simplified" License
177 stars 68 forks source link

Sort out which version of BEL to support #108

Closed bgyori closed 7 years ago

bgyori commented 8 years ago

We currently have inconsistent BEL namespaces in the different files we use. For instance, the latest BEL framework uses SFAM as the namespace for protein families and updates the names of some of the entries: http://resource.belframework.org/belframework/latest-release/ However, the BEL Large Corpus hosted on NDEx, which we access via the indra.bel.process_ndex_neighborhood API uses the old BEL framework: http://resource.belframework.org/belframework/1.0/ This means that, for instance, human protein families are defined in the PFH name space. Meanwhile, our offline BEL files currently have inconsistent standards. For instance, https://github.com/sorgerlab/indra/blob/master/data/small_corpus.rdf uses the new standard but https://github.com/sorgerlab/indra/blob/master/data/RAS_neighborhood_extended.rdf uses the old standard.

bgyori commented 8 years ago

Another inconsistency we need to deal with is that the NDEx web service refers to entities as http://www.openbel.org/bel/namespace//MAPK1 with two forward slashes before the entity name, whereas all the offline files in indra/data use http://www.openbel.org/bel/namespace/MAPK1

bgyori commented 7 years ago

These are now solved and the BEL processor supports both formats to the extent possible.