A novel approach for predicting upstream regulators (PURE) that affect gene expression.

Area of Science:

• Genomics
• Toxicology
• Pharmacology

Background:

• External factors like chemical, drug, or toxicant (CDT) exposure can cause diseases.
• Identifying causal CDTs from gene expression changes is vital for research and drug discovery.
• Inferring CDTs that reverse disease-induced gene expression is key for drug repurposing.

Purpose of the Study:

• To introduce the Predicting Upstream REgulators (PURE) approach for identifying causal CDTs.
• To evaluate PURE's ability to infer CDTs from gene expression data.
• To assess PURE's utility in identifying drugs for reversing disease phenotypes.

Main Methods:

• PURE was developed to infer causal CDTs from gene expression profiles.
• The approach was compared against four classical methods and Ingenuity Pathway Analysis (IPA).
• Performance was evaluated on 16 datasets (rat, mouse, human) involving 8 chemicals/drugs, assessing CDT identification rank and false positives.

Main Results:

• PURE outperformed existing methods in 11 out of 16 experiments.
• PURE correctly identified the causal CDT at the top rank 7 times.
• IPA was second-best but failed to identify the correct CDT in 5 experiments.

Conclusions:

• PURE effectively infers true CDTs responsible for observed gene expression changes.
• The PURE approach demonstrates superior performance compared to popular methods.
• PURE shows significant potential for applications in drug repurposing and understanding disease etiology.