Mapping chemical-gene interactions for developmental lethality and pregnancy loss

Purpose: Pregnancy loss affects 10-15% of clinically recognized pregnancies and is driven by genetic and environmental factors. Interactions between genes and chemicals critically shape developmental outcomes, yet existing resources do not organize chemical-gene evidence by gestational timing, maternal-fetal compartment, or lethality context. Methods: To minimize this gap, we merged Intolerome genes and the Comparative Toxicogenomics Database databases to create a filtered network of 928 lethality-associated genes and ~4,000 chemicals. We developed the Chemical-Gene Atlas (CGA) application, which analyzes chemical-gene interactions using four filters supporting hypothesis generation and clinical translation. Results: Using recurrent pregnancy loss (RPL) as a case study, CGA identified five clinically important genes (F5, F2, AURKB, PADI6, and FOXD1) demonstrating different exposomic patterns with Bisphenol A (BPA) and Benzo[a]pyrene (B[a]P). These exposures affect gene expression, methylation, and coagulation pathways and placental function. Our findings demonstrate that genes with loss-of-function intolerance, especially those expressed in metabolic and cardiovascular systems, are susceptible to environmental disruption during critical windows of development, including implantation, placental development, and early organogenesis. Conclusion: The CGA platform provides a scalable model to investigate chemical-gene interactions leading to developmental lethality. CGA is available at https://cgatlas.org/.