Using fragment chemistry data mining and probabilistic neural networks in screening chemicals for acute toxicity to the fathead minnow
Bibliographic record
Abstract
The paper is illustrating how the general data mining methodology may be adapted to provide solutions to the problem of high throughput virtual screening of organic chemicals for possible acute toxicity to the fathead minnow fish. The present approach involves mining fragment information from chemical structures and is using probabilistic neural networks to model the relationship between structure and toxicity. Probabilistic neural networks implement a special class of multivariate non-linear Bayesian statistical models. The mathematical principles supporting their use for value prediction purposes are clarified and their peculiarities discussed. As part of the research phase of the data mining process, a dataset consisting of 800 structures and associated fathead minnow (Pimephales promelas) 96-h LC50 acute toxicity endpoint information is used for both the purpose of identifying an advantageous combination of fragment descriptors and for training the neural networks. As a result, two powerful models are generated. Model 1 implements the basic PNN with Gaussian kernel (statistical corrections included) while Model 2 implements the PNN with Gaussian kernel and separated variables. External validation is performed using a separate dataset consisting of 86 structures and associated toxicity information. Both learning and generalization capabilities of the two models are investigated and their limitations discussed.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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 source (direct Gemma or distilled Codex), 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".