PandaOmics Data Processing Pipeline (IMAGE)
Caption
PandaOmics employs a robust data processing pipeline, starting with dataset selection (1) and sample group comparison (2). It offers gene- and pathway-level analysis (3), and combines the results of these steps into meta-analyses (4), enhancing target predictions with diverse data sources including prior knowledge extracted from text data (5). For therapeutic target and biomarker identification (6), it provides a user-friendly interface with 23 disease-specific models. Identification of disease-relevant compounds is also enabled (7). PandaOmics excels in indication prioritization and features a database of precalculated disease meta-analyses (8). It also leverages biological knowledge graph (9) and large language model-based ChatPandaGPT (10) to explain gene-disease associations. Robotic lab (11) for target validation and compound screening forms a feedback loop with the AI core of the platform, streamlining the research process. The target hypotheses identified with PandaOmics serve as an input for Chemistry42 software (12) to perform small molecule generation.
Credit
Insilico Medicine
Usage Restrictions
none
License
Original content