Development of an adverse outcome pathway for radiation-induced microcephaly via expert consultation and machine learning
Bibliographic record
Abstract
Background Brain development during embryogenesis and in early postnatal life is particularly complex and involves the interplay of many cellular processes and molecular mechanisms, making it extremely vulnerable to exogenous insults, including ionizing radiation (IR). Microcephaly is one of the most frequent neurodevelopmental abnormalities that is characterized by small brain size, and is often associated with intellectual deficiency. Decades of research span from epidemiological data on in utero exposure of the A-bomb survivors, to studies on animal and cellular models that allowed deciphering the most prominent molecular mechanisms leading to microcephaly. The Adverse Outcome Pathway (AOP) framework is used to organize, evaluate and portray the scientific knowledge of toxicological effects spanning different biological levels of organizations, from the initial interaction with molecular targets to the occurrence of a disease or adversity. In the present study, the framework was used in an attempt to organize the current scientific knowledge on microcephaly progression in the context of ionizing radiation (IR) exposure. This work was performed by a group of experts formed during a recent workshop organized jointly by the Multidisciplinary European Low Dose Initiative (MELODI) and the European Radioecology Alliance (ALLIANCE) associations to present the AOP approach and tools. Here we report on the development of a putative AOP for congenital microcephaly resulting from IR exposure based on discussions of the working group and we emphasize the use of a novel machine-learning approach to assist in the screening of the available literature to develop AOPs.Conclusion The expert consultation led to the identification of crucial biological events for the progression of microcephaly upon exposure to IR, and highlighted current knowledge gaps. The machine learning approach was successfully used to screen the existing knowledge and helped to rapidly screen the body of evidence and in particular the epidemiological data. This systematic review approach also ensured that the analysis was sufficiently comprehensive to identify the most relevant data and facilitate rapid and consistent AOP development. We anticipate that as machine learning approaches become more user-friendly through easy-to-use web interface, this would allow AOP development to become more efficient and less time consuming.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".