In chemical safety and toxicology, consistency matters. When scientists assess whether a new fragrance ingredient or chemical compound is safe, they often rely on data from related substances that have already been tested. This approach—known as read-across—is a cornerstone of modern safety assessment and plays a vital role in reducing animal testing.
But read-across depends on one surprisingly difficult question:
What does it really mean for two chemicals to be “similar”?
For years, the answer has been unclear. Grouping decisions often relied on expert judgment, simple similarity scores, or a mix of both. While these approaches can be helpful, they are not always transparent or consistent. To address this, the Research Institute for Fragrance Materials (RIFM) has introduced the RIFM Grouping Method—a structured way to define chemical similarity using a clear, standardised structural “signature.”
Rather than assigning a single similarity score, the method describes chemical structure in a way that is both reproducible and easy to explain.
If you’re new to chemical safety, grouping methods, or read-across, here’s a simple way to think about it. Testing every new chemical from scratch is time-consuming and often unnecessary. Instead, scientists compare new substances to similar ones that have already been studied. The challenge is making sure those comparisons are fair, consistent, and scientifically sound. That’s where structured grouping methods—like the one developed by RIFM—come in.
Why Chemical Grouping Has Been So Challenging
Before looking at the solution, it helps to understand why traditional grouping methods often struggle.
Subjective expert judgment
Expert insight is essential in toxicology, but when grouping relies mainly on individual interpretation, results can vary. One assessor may see Chemical A as a good match for Chemical B, while another may disagree. This fact makes it harder to reproduce decisions and justify them to regulators.
Over-simplified similarity scores
Tools like the Tanimoto index compare chemical structures mathematically, but they don’t always reflect how substances behave in the body. Two molecules can look very similar on paper and still differ in metabolism, reactivity, or toxicity. High similarity scores can therefore give a false sense of confidence.
What was missing was an approach that is consistent, transparent, and biologically meaningful—not just computationally efficient.
A Different Approach: Structure–Activity Groups (SAGs)
The RIFM Grouping Method takes a different route. Instead of relying on abstract scores, it uses a structured “language” to describe chemical features. The key concept is the Structure–Activity Group (SAG).
The system is built in three layers:
Indicator Phrases
Short, standardised labels that describe specific structural features, such as aldehyde, branched-chain, or aromatic.
Chemical Signature
A structured string of indicator phrases that together describe the whole molecule, ordered from the most essential feature to the least.
Structure–Activity Group
A group of chemicals that share the same chemical signature.
You can think of the chemical signature as a structural fingerprint in words—precise enough for computers, but still readable by scientists.
How a Chemical Signature Is Built
RIFM illustrates the process using 3-phenylbutanal (CAS RN 16251-77-7).
Step 1: Identify key heteroatoms
The process starts by identifying non-carbon atoms that strongly influence chemical behaviour.
- In this case, oxygen is present
- Result: oxygen_containing
Step 2: Identify the primary functional group
Using a predefined priority order, the most relevant functional group is selected.
- For 3-phenylbutanal, the aldehyde group is the key feature
- Result: /aldehyde/
Step 3: Describe the carbon framework
The remaining structure is described by answering a series of simple questions:
- Is the structure linear or cyclic?
- Is it branched?
- Is it saturated?
- Are aromatic rings present?
- What is the carbon chain length?
- For 3-phenylbutanal, this leads to:
/alkyl_chain_cyclic/branched_chain/saturated/monocyclic/aromatic/c6-c13/
The final signature
Putting everything together gives the full chemical signature:
/oxygen_containing/aldehyde/alkyl_chain_cyclic/branched_chain/saturated/monocyclic/aromatic/c6-c13/
The result is a clear, unambiguous description of the molecule that can be reproduced every time.
What This Means for Read-Across
The real value of Structure–Activity Groups becomes clear when they are used for read-across, where safety data from a tested chemical is applied to an untested one.
Instead of assuming similarity, the RIFM method makes it visible by grouping chemicals into clear tiers.
Tier I: Same signature
Example:
- Target: Amyl Alcohol (CAS RN 71-41-0)
- Source: Propyl Alcohol (CAS RN 71-23-8)
These chemicals share the same SAG, which provides the highest confidence for read-across.
Tier II: One clear difference
Example:
- Target: Amyl Alcohol (CAS RN 71-41-0)
- Source: Isoamyl Alcohol (CAS RN 123-51-3)
Here, the signatures differ by just one indicator phrase, such as a branched versus a straight chain. Because that difference is clearly identified, scientists can focus on whether it matters for toxicity—rather than debating overall similarity.
This kind of transparency makes read-across decisions easier to explain and defend.
Why This Approach Matters
The impact of the RIFM Grouping Method goes beyond better organisation.
- Consistency
Chemicals are grouped the same way every time, regardless of who performs the assessment. - Transparency
The logic behind each grouping decision is clear and easy to communicate. - Reduced bias
The method limits subjective choices while still allowing expert interpretation where needed. - Proven use
Between 2020 and 2022, the approach was applied to more than 900 chemical pairs, enabling more efficient assessment of analog suitability and highlighting meaningful differences.
Most importantly, it supports stronger, more defensible read-across justifications, which are increasingly crucial for regulatory acceptance and non-animal testing strategies.
From “Similar Enough” to “Clearly Defined”
The RIFM Grouping Method marks a shift in how chemical similarity is handled—from informal judgment to clearly defined structure-based reasoning. By turning similarity into something that can be shown rather than assumed, it strengthens both scientific confidence and regulatory credibility.
For organisations looking to reduce testing while maintaining robust safety standards, this kind of clarity is essential. Well-designed read-across is not just about having data—it’s about being able to explain, justify, and stand behind the decisions made.
