Biology subjects

Paparella, M.

Publications and source records attributed to Paparella, M..

4 recordsLinked to original sources

bioRxiv2026

Integrating Semantic Retrieval, LLM-based Refinement, and Structured Expert Curation for Scalable AOP Gene Mapping

Toxicogenomics can support regulatory toxicology, but its use is limited by the difficulty of translating molecular responses into mechanistic, decision-relevant interpretations. Adverse Outcome Pathways (AOPs) provide a framework for this translation, yet omics applications require scalable mapping of Key Events (KEs) to molecular features. Here, we present an AI-assisted, multi-step workflow for KE-to-gene mapping that uses embedding-based semantic retrieval to identify candidate ontology/pathway terms, large language model-assisted refinement to filter these candidates, and double-independent expert group curation with rule-based consolidation to finalize mappings and derive confidence scores. Compared with earlier NLP-based approaches, the workflow improves KE-to-ontology/pathway mapping performance and generates candidate annotations that better align with expert judgment while substantially reducing the need for manual augmentation. Explicit gene and protein mentions in KE titles were additionally grounded to improve specificity, and each curated mapping was assigned curator reason codes to support transparent, traceable, and confidence-aware reuse. Applied across AOP-Wiki, the workflow produced a comprehensive KE-to-gene set resource covering 1,254 KEs across 523 AOPs and linking 15,833 human genes. Utility is demonstrated through CTD-based AOP fingerprinting of curated reference chemical groups, highlighting expanded coverage and confidence-informed interpretation of chemical-associated gene signatures in an AOP context. The workflow and resulting resource provide a practical bridge between toxicogenomics and AOP-based mechanistic interpretation and support routine updating and future extension to additional omics layers within OECD Omics2AOP.

bioinformatics↗

bioRxiv2025

A Data-Driven Approach for the Development of a Time-informed Adverse Outcome Pathway-network for Cardiotoxicity of Environmental Chemicals

We present a novel Adverse Outcome Pathway (AOP) network for environmental chemical-induced cardiotoxicity using a bottom-up, data-driven AOP development approach. Mechanistic endpoints were systematically extracted from 339 in vitro and in vivo studies, yielding 1,759 Key Event (KE) entries and 4,938 Key Event Relationship (KER) entries, including information on experimental methods, essentiality evidence (intervention experiments demonstrating upstream-downstream dependence), and study metadata. After quality filtering (high risk of bias, confounding cytotoxicity in vitro, excessive toxicity or animal well-being concerns in vivo, and low-frequency observations), 112 unique KEs and 829 unique KERs supported by at least three independent observations were retained for network construction. Network analysis identified oxidative stress and mitochondrial dysfunction as dominant hub processes linking diverse upstream perturbations to downstream cardiomyocyte injury, inflammation, cardiac remodelling (fibrosis and hypertrophy), decreased cardiac contractility, and reduced left ventricular function. Incorporating exposure duration at the KER level enabled time-resolved pathway interpretation and demonstrated that KE timing is relationship-dependent, revealing temporal patterns not apparent when analysing KEs in isolation. This evidence-weighted, time-resolved AOP network can support endpoint prioritisation and exposure-window selection for non-animal method (NAM) test batteries and mechanistically informed cardiotoxicity assessment. SynopsisEnvironmental chemicals converge on shared stress and injury pathways that drive cardiac remodelling and ventricular dysfunction. A time-resolved AOP network helps prioritise endpoints and exposure windows for non-animal cardiotoxicity testing.

pharmacology and toxicology↗

bioRxiv2025

Variability and uncertainty of data from genotoxicity Test Guidelines: What we know and why it matters.

This review comprehensively examines the variability and uncertainty associated with test guideline (TG)-conform genotoxicity data and explores the respective implications for the integration of non-animal-methods (NAMs) into regulatory frameworks. Historical amendments to OECD TGs are mapped to reveal the methods evolution that improves the scientific quality of the data but also explains data heterogeneity within available databases. An analysis of the major genotoxicity databases ECVAM, ISSMIC, and OASIS demonstrates substantial variability in genotoxicity calls. Using the EFSA genotoxicity database, which currently harbours the best-curated (meta-) data, we estimate that 22-77% of compounds exhibit similarity of replicate results below 85%, depending on the assay. The potentially most important variables statistically explaining variability and sensitivity were analysed. The practical limitations to identify them with high reliability and to define their optimum needs to be accepted as a qualitative baseline uncertainty. These findings underscore the necessity of contextualizing NAM performance evaluations within the intrinsic variability and uncertainty of animal and in vitro reference data. We propose that this variability is explicitly considered in the development and validation of NAM-based Integrated Approaches for Testing and Assessment (IATAs). This review provides a critical foundation for regulators and scientists aiming to enhance the acceptance and utility of NAMs in genotoxicity assessment.

pharmacology and toxicology↗

bioRxiv2024

Daphnids Can Safeguard the Use of Alternative Bioassays to the Acute Fish Toxicity Test: A Focus on Neurotoxicity

Assessment of potential impacts of chemicals on the environment traditionally involves regulatory standard data requirements for acute aquatic toxicity testing using algae, daphnids and fish (e.g., OECD test guidelines (TG) 201, 202, and 203, respectively), representing different trophic levels. In line with the societal goal to replace or reduce vertebrate animal testing, alternative bioassays were developed to replace testing with fish: the fish cell line RTgill-W1 acute toxicity assay (OECD TG249) and the zebrafish embryo acute toxicity test (zFET, OECD TG236). However, previous studies revealed the lower sensitivity of the RTgill-W1 cell line assay and zFET for some neurotoxic chemicals and allyl alcohol, which is presumably biotransformed in fish to the more toxic acrolein (which is predicted well through the cell line assay). To provide an additional alternative to acute fish toxicity, in this study, we analyzed historic ecotoxicity data for fish and daphnids from the EnviroTox Database. We found a considerable variability in acute fish LC50 and acute daphnids EC50 values, particularly for neurotoxic chemicals. Comparing sensitivity of these taxonomic groups according to different neurotoxicity classification schemes indicates that fish rarely represent the most sensitive trophic level of the two. Exceptions here most prominently include a few cyclodiene compounds, which are no longer marketed, and a chemical group that could be identified through structural alerts. Moreover, daphnids are more sensitive than fish to acrolein. This analysis highlights the potential of the Daphnia acute toxicity test, which is usually a standard regulatory data requirement, in safeguarding the environmental protection level provided by the RTgill-W1 cell line assay and the zFET. This research, rooted in decades of efforts to replace the fish acute toxicity test, shifts the focus from predicting fish toxicity 1-to-1 to emphasizing the protectiveness of alternative methods, paving the way for further eliminating vertebrate tests in environmental toxicology.

pharmacology and toxicology↗