Related Experiment Video
Updated: Jun 30, 2025
08:21
Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids
Published on: April 13, 2022
2.6K
Systematic Approaches for the Encoding of Chemical Groups: A Case Study.
Panagiotis G Karamertzanis1, Grace Patlewicz2, Marta Sannicola1
1Computational Assessment and Alternative Methods, European Chemicals Agency (ECHA), Telakkakatu 6, Helsinki 00150, Finland.
Chemical Research in Toxicology
|March 18, 2024
Summary
Machine learning models can now preliminarily assign chemical substances to predefined groups, improving upon manual curation. Random forest models showed higher accuracy than k-nearest neighbor for this task in chemical risk assessment.
Area of Science:
- Computational toxicology
- Cheminformatics
- Machine learning in regulatory science
Background:
- Regulatory agencies group chemicals for hazard and risk assessment, but manual curation is challenging to update.
- Existing chemical groupings lack explicit structural or property-based rules, hindering efficient updates.
- Manual expert curation of chemical groups is time-consuming and difficult to refine with new data.
Purpose of the Study:
- To develop machine learning models for preliminary assignment of substances to predefined groups.
- To automate and streamline the initial profiling of new chemical substances for regulatory assessment.
- To improve the efficiency of prioritization in hazard and risk assessment processes.
Main Methods:
- Mapped 86 European Chemicals Agency (ECHA) groupings to U.S. Environmental Protection Agency (EPA) DSSTox database.
- Utilized Morgan fingerprints for chemical and structural representation of substances.
- Employed k-nearest neighbor (kNN) and random forest (RF) machine learning classifiers for group assignment.
Main Results:
- Random forest (RF) achieved a mean 5-fold cross-validation accuracy (F1 score) of 0.853, outperforming kNN (0.781).
- RF classifier demonstrated a statistically significant 9% improvement in accuracy over kNN (p-value = 0.001).
- Models successfully classified substances into 56 groups with at least 10 members and structural data.
Conclusions:
- Machine learning, particularly RF, offers a promising approach for initial substance profiling into predefined groups.
- This method can facilitate prioritization and streamline the assessment of new substances.
- The developed algorithm is publicly available, enabling model use and refitting with new data.
More Related Videos
09:04Identifying Per- and Polyfluorinated Chemical Species with a Combined Targeted and Non-Targeted-Screening High-Resolution Mass Spectrometry Workflow
Published on: April 18, 2019
12.4K
08:09Annotation of Plant Gene Function via Combined Genomics, Metabolomics and Informatics
Published on: June 17, 2012
19.7K
Related Concept Videos
02:08Introduction to Functional Groups
26.5K
Functional groups are group of atoms with specific chemical properties that occur within organic molecules and sometimes denoted as “R”. Functional groups are found along the carbon backbone of macromolecules can form chains or rings of carbon atoms. Functional groups can “functionalize” a compound by enabling it to adopt different physical and chemical properties.
Types of common functional groups
The table below summarizes some of the major functional...
26.5K
02:22Overview of Advanced Functional Groups
23.9K
Functional groups are groups of atoms with specific chemical properties that occur within organic molecules and are sometimes denoted as “R”. Functional groups can “functionalize” a compound by enabling it to adopt different physical and chemical properties.
Types of Advanced Functional Groups
The table below summarizes some of the major functional groups in organic chemistry.
23.9K
01:09¹H NMR: Pople Notation
1.8K
The Pople nomenclature system classifies spin systems based on the difference between their chemical shifts. Coupled spins are denoted by capital letters with subscripts indicating the number of equivalent nuclei. When the coupled nuclei have well-separated chemical shifts, they are assigned letters that are far apart in the alphabet, such as A and X. When the difference in chemical shifts is small, coupled nuclei are named using adjacent letters of the alphabet (AB, MN, or XY).
A proton...
A proton...
1.8K
01:19Overview of Functional Groups
10.8K
Functional groups are a group of atoms with characteristic properties, which when linked to the carbon skeleton of a molecule, alter the properties of that molecule. For example, certain functional groups will make a molecule hydrophilic, whereas others will make them hydrophobic. These functional groups are an indispensable part of organic chemistry and important components of biological molecules, such as carbohydrates, proteins, lipids, and nucleic acids. Each functional group is a unique...
10.8K
02:05Prochirality
3.8K
The concept of prochirality leads to the nomenclature of the individual faces of a molecule and plays a crucial role in the enantioselective reaction. It is a concept where two or more achiral molecules react to produce chiral products. A typical process is the reaction of an achiral ketone to generate a chiral alcohol. Here, the achiral reactant reacts with an achiral reducing agent, sodium borohydride, to generate an equimolar mixture of the chiral enantiomers of the product. For example, an...
3.8K
01:23Protecting Groups for Aldehydes and Ketones: Introduction
7.0K
Protecting groups are compounds that can bind to a specific functional group in the presence of other functional groups to protect them from undesired chemical reactions. These compounds can selectively bind to particular functional groups and advance chemoselective reactions in polyfunctional systems (Figure 1). After the functional group has served its purpose, it is removed by reacting it with specific compounds.
7.0K