QSAR approaches to predicting toxicity
Abstract
Due to the demands of time and the high cost of testing compounds for toxicity in test animals, it would be an advantage to be able to estimate the toxic response of chemical agents using theoretical approaches. Predicting whether a compound will be toxic or nontoxic is a classification problem and the methods of studying quantitative structure activity relationships (QSAR) can be used for this purpose [Hansch, C. (1969) Accounts Chem. Res., 2, 232]. It should be recognized, however, that formulating the QSAR problem as one of active vs. inactive makes it different from classical QSAR problems. This requires that methods be applied that can predict the category of a compound to be used, i.e., so-called methods of pattern recognition (Varmuza, K. (1983) J. Chem. Info. Comp. Sci. 23, 6) being required. There are several methods of pattern recognition that can be used with some being more suitable than others. The nature of this unique QSAR problem, the appropriate methods to apply, and some of the pitfalls of applying QSAR techniques to predicting toxicity are discussed.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Simple pattern recognition is difficult to process asymmetric data, in which positive samples are far less than negative ones, or vice versa.
“simple pattern recognition is difficult to process asymmetric data, in which positive samples are far less than negative ones, or vice versa [109].”
From Machine Learning Based Toxicity Prediction: From Chemical Structural Description to Transcriptome Analysis (2018) · cites this paper as [109]
Cites
Pattern Recognition in Chemistry
Вармуза, К. (1980) · Lecture notes in chemistry · doi:10.1007/978-3-642-93155-0 · Cited by 237 · see citing works
Four levels of pattern recognition
Albano, C., Dunn, W., Edlund, U. G., et al. (1978) · Analytica Chimica Acta · doi:10.1016/s0003-2670(01)83107-x · Cited by 172 · see citing works
Self-citationWold, S., & Dunn, W. J. (1983) · Journal of Chemical Information and Computer Sciences · doi:10.1021/ci00037a002 · Cited by 165 · see citing works
Self-citation31 Pattern recognition in chemistry
Kowalski, B. R., & Wold, S. (1982) · Handbook of statistics · doi:10.1016/s0169-7161(82)02034-3 · Cited by 70 · see citing works
Structure-activity analyzed by pattern recognition: the asymmetric case
Dunn, W. J., & Wold, S. (1980) · Journal of Medicinal Chemistry · doi:10.1021/jm00180a003 · Cited by 58 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Wu, Y., & Wang, G. (2018) · International Journal of Molecular Sciences · doi:10.3390/ijms19082358 · Cited by 193 · see citing works
- Authors
- Dunn, W. J.
- Affiliations
- University of Illinois Chicago (US)
- Corresponding author
- Dunn, W.J.
- Venue
- Toxicology Letters
- Publisher
- Elsevier BV
- Published
- 1988-10
- Volume
- 43
- Issue
- 1-3
- Pages
- 277-283
- ISSN
- 0378-4274 (ISSN-L)
- OpenAlex
- W1995240112
- Cited by
- 21
- References
- 5
- Citations by year
- 2026: 1 · 2025: 1 · 2024: 1 · 2023: 3 · 2022: 1 · 2020: 1 · 2019: 3 · 2018: 1 · 2017: 1 · 2016: 1
- FWCI
- 0.49
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Toxicology Letters (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
- Health Sciences › Veterinary › Small Animals › Animal testing and alternatives
- Physical Sciences › Chemistry › Physical and Theoretical Chemistry › History and advancements in chemistry
- Keywords
- QSAR, toxicity prediction, pattern recognition, alternatives to animal testing, chemical toxicity
- MeSH terms
- Computer Simulation, Structure-Activity Relationship, Toxicology
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
