/phi-llemma

Tools

  • BioGRID
  • NCBIEntrez
  • Pubmed
  • NASA API
  • BioOnto
  • SPARQL
  • BioPython

Datasets

  • ProtocolsIO
  • PubmedCentral
  • Linked SPARQL Queries

Knowledge-Augmented Pretraining Workflow

  1. Gather PMIDs/ISBNs from specialist datasets (microbe x microbe, microbe x environment, BioGRID)
  2. Retrieve/generate tags for each abstract

Questions

  • Title Generation
  • NER (pubtator)
  • Genes
  • Proteins
  • Organisms
  • Diseases
  • Chemicals
  • Cell lines