INDRA v1.19.0

Package structure and dependencies

• Setup adjusted to require newer versions of PyBEL (0.15), PyBioPax (0.0.5), and Protmapper (0.0.19).
• Automated testing has been migrated from Travis CI to Github Actions.

Core assembly modules

• The statement refinement finding algorithm has been reimplemented for a speedup of multiple orders of magnitude depending on application in indra.preassembler.
• The Preassembler now supports a series of user-supplied filter functions to speed up refinement finding in a use case-specific way in indra.preassembler.
• The BioOntology's version has increased to 1.7 with many resource updates and new sources (e.g., Biomappings included in indra.ontology.
• The IndraOntology no longer contains connected component labels (faster initialization and loading), in indra.ontology.
• The grounding disambiguation code is refactored with improved prioritization in indra.preassembler.grounding_mapper.
• Grounding disambiguation has been refactored into a separate module in indra.preassembler.grounding_mapper.disambiguate.
• New Statement validation module for automated consistency checks of Statements, Agents, and their attributes in indra.statements.validate.
• Improvements to WorldOntology extension during runtime in indra.ontology.world.

Knowledge sources

• Structured meta-information on INDRA's sources is now available as a resource file in indra/resources/source_info.json.
• The deprecated LINCS Small Molecules (indra.sources.lincs_drug) source module has been removed; replaced by TAS in indra.sources.tas.
• The BEL processor is now compatible with PyBEL 0.15+ in indra.sources.bel.
• Updaed BioGRID processor, more extractions and compatibility with the TAB 3.0 format in indra.sources.biogrid.
• Improvements to BioPAX processing to support e.g., BioPAX from Reactome in indra.sources.biopax.
• Support for compositional grounding from the CWMS, Eidos, Sofia, and Hume readers.
• Support for event type filter for the CWMS, Eidos, Sofia, and Hume readers.
• Grounding cache feature added for re-grounding in indra.sources.eidos.
• Support for using obtained evidence counts in INDRA DB REST client and HTML assembler.
• Support for specifying organism priority for protein grounding in Reach processor in indra.sources.reach.
• Handle protein chain/fragment grounding in Reach processor in indra.sources.reach.
• Improvements to extracting some forms of regulation in Reach processor in indra.sources.reach
• Many improvements to Signor processor in indra.sources.signor.
• New EKB forms extracted in TRIPS processor indra.sources.trips.

Model assemblers

• New HTML look implemented (with support for old/simplified view) in the HTML Assembler in indra.assemblers.html.

Model analysis / explanation

• Many new path finding algorithms implemented in indra.explanation.pathfinding.
• Support for semantic filters on paths implemented for model checking in indra.explanation.model_checker.

Resources and database clients

• indra.resources is now a proper module with helper functions to load resource files.
• New client for interacting with identifiers.org namespaces, IDs and URLs in indra.databases.identifiers.
• Many resources have been updated (e.g., HGNC entries, FamPlex, new grounding mappings) in indra.resources.

Literature clients

• Better MeSH annotation extraction, including supplementary concepts in PubMed client indra.literature.pubmed_client.
• Local storage feature for DART client in indra.literature.dart_client.