Drug Discovery AI Workflow Construction
Provides solutions for the three core aspects of drug discovery and development:
- Literature Mining::
- utilizationOpenMed-NER-ChemicalSeries extracts compounds from PubMed literature
- add sth. into a groupRelation ExtractionModeling Role Relationship Networks
- Example: Drug-target pairs from "Ridecivir inhibits SARS-CoV-2 protease activity".
- Analysis of test data::
- Processing of electronic laboratory records (ELN) was performed with theOpenMed-NER-DosageIdentify concentration data
- Physical linkage technology to standardize "5-FU" to "Fluorouracil"
- knowledge graph construction::
from openmed import build_kg kg = build_kg(research_papers, entity_types=['CHEMICAL','GENE','DISEASE'], model='OpenMed-NER-MultiDetect-434M')
The Pfizer team's case shows that after adopting OpenMed's compound identification process, the efficiency of literature screening increased by 40% and the new drug target discovery cycle was shortened from 6 months to 3 months. The key is integration in the pipelineOpenMed/OpenMed-NER-ADMET-290MModeling to predict compound properties.
This answer comes from the articleOpenMed: an open source platform for free AI models in healthcareThe
May not be reproduced without permission:AI productivity tools " How can OpenMed be applied to accelerate compound screening in drug discovery scenarios?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil