An optimized gene set for transcriptomics based neurodevelopmental toxicity prediction in the neural embryonic stem...

Area of Science:

• Toxicology
• Developmental Biology
• Genomics

Background:

• The neural embryonic stem cell test (ESTn) is an in vitro model for assessing neurodevelopmental toxicity.
• Transcriptomics analysis in ESTn offers improved prediction accuracy and mechanistic insights.
• Distinguishing between neural differentiation and toxicant exposure is crucial for accurate ESTn interpretation.

Purpose of the Study:

• To identify an optimized gene set for neurodevelopmental toxicity prediction within the ESTn.
• To enhance the reliability and accuracy of ESTn-based toxicity assessments.
• To refine transcriptomics analysis for improved compound risk evaluation.

Main Methods:

• Performed de novo analysis on combined raw transcriptomics data from 10 compounds and 19 exposures.
• Evaluated 200,000 randomly selected gene sets to identify predictive genes.
• Applied Principal Component Analysis-based differentiation track algorithm to distinguish biological responses.

Main Results:

• Identified a 100-gene set predominantly involved in neural development, significantly contributing to prediction reliability.
• Further refined the gene set to 29 genes, achieving 84% prediction accuracy (Area Under the Curve 94%).
• Demonstrated the effectiveness of optimized gene sets in predicting neurodevelopmental toxicity.

Conclusions:

• The identified 29-gene set significantly improves prediction accuracy in the ESTn.
• These optimized gene sets are valuable for advancing transcriptomics-based compound risk assessment.
• Further application of these gene sets is anticipated to enhance ESTn utility in toxicology studies.