Predicting the toxicity of chemical compounds via Hyperdimensional Computing
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
ABSTRACT Accurately and efficiently assessing the potential toxicity of chemical compounds is critical given their wide application across pharmaceutical, industrial, and environmental domains. Traditional toxicological evaluations, which predominantly rely on intensive in vitro and in vivo assays, are frequently slow and expensive processes. Here, we introduce a novel application of Hyperdimensional Computing (HDC), an emerging computational paradigm inspired by the way the human brain works in encoding information, for the efficient classification of chemical compounds as either toxic or non-toxic. Our methodology employs Simplified Molecular Input Line Entry System (SMILES) representations of compounds, drawing data from the comprehensive Tox21 dataset. We delineate a pipeline wherein these chemical structures are encoded into high-dimensional binary vectors, which subsequently serve as the foundation for training and classification within the HDC framework. This approach leverages HDC’s inherent advantages, including its resilience to noise, parallel processing capabilities, and efficacy in identifying intricate patterns. This work demonstrates the viability of HDC as a promising alternative for large-scale toxicity prediction, offering a computationally efficient and scalable solution. This research significantly contributes to the field of cheminformatics by validating HDC’s potential in chemical property prediction, thereby facilitating accelerated identification of hazardous substances and mitigating the reliance on intensive laboratory experimentations.
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.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 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 it