INDRA v1.20.0

Package structure and dependencies

• INDRA is now tested with Python 3.9 (supports 3.6-3.9).
• Increased protmapper version (0.0.21) with updated protein resources.
• Moved out deprecated evaluation scripts (#1283).

Core assembly modules

• Preassembler refactored to support refinement filter classes to enable incremental assembly indra.preassembler.
• Belief engine refactored to support incremental assembly indra.belief.
• New function to get standard Agent from name and references in indra.ontology.standardize.
• Support for intermediate node annotations and local file parsing in YAML ontology indra.ontology.world.
• INDRA BioOntology updated from v1.7 to v1.12 with bug fixes and resource updates indra.ontology.bio.
• New function to re-map DB refs within the same namespace in indra.tools.assemble_corpus.

Statement representation

• New pretty print and statement statistics functions in indra.statements.io.
• Fixes related to refinement finding and hash calculation in indra.statements.
• Fixed refinement finding for multiple agents with the same name in bound conditions indra.statements.agent.

Knowledge sources

• New API and processor for the Drug Gene Interaction (DGI) database at indra.sources.dgi.
• New API and processor for the Chemical Roles Graph (CRoG) at indra.sources.crog.
• New API and processor for MINERVA models at indra.sources.minerva.
• New function to process a single BEL statement string in indra.sources.bel.
• Improved chemical grounding in BioPAX processor indra.sources.biopax.
• New process from web function for DrugBank indra.sources.drugbank.
• EidosBioProcessor refactored into a separate class indra.sources.eidos.
• Eidos processor allows using custom grounder when processing into bio statements indra.sources.eidos.bio_processor.
• New feature to turn Statements into Hypothes.is annotations and upload them in indra.sources.hypothesis.
• Support for processing BEL script annotations from Hypothes.is in indra.sources.hypothesis.
• New processor to process remote INDRA JSON files indra.sources.utils.

Model assemblers

• Statement presentation refactored to support custom grouping and sorting indra.util.statement_presentation.
• HTML assembler builds on new statement presentation to support custom sorting indra.assemblers.html.
• IndraNet assembler supports adding extra columns/properties to the output indra.assemblers.indranet.
• PySB assembler allows reverse-engineering Agents from an assembled model indra.assemblers.pysb.

Model analysis / explanation

• INDRA is now compatible with Kappy 4.1.2 (which changes the contact map format).
• New Kappa contact map analysis to diagnose cycles in indra.assemblers.pysb.kappa_util.
• New NodeContainer class for model checking to carry node annotations in indra.explanation.model_checker.
• Support for custom edge filters in path finding in indra.explanation.path_finding.
• Support for reporting refinement edges in path finding in indra.explanation.reporting.

Tools

• New tool to upload annotations to Hypothes.is from INDRA Statements in indra.tools.hypothesis_annotator.

Resources and database clients

• Resource updates to Biomappings, ChEBI and EFO resources.
• Many new grounding curations added to grounding resources (mappings, misgrounding, ignores).
• Better handling for external namespaces in the OBO client indra.databases.obo_client.

Literature clients

• The DART client was moved into the INDRA World package.
• New function to get titles for PMC articles in indra.literature.pmc_client.