iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
A 2023 study used multiple molecular fingerprints and reaction-mapping tools to explore how FDA-approved drugs resemble one another—and how complex the structural changes between selected pairs appear. Its maps can help researchers generate hypotheses, but they do not show that two drugs are interchangeable or that one can practically be synthesized from the other.
What the study set out to show
In “Alchemical analysis of FDA approved drugs,” published in Digital Discovery on August 30, 2023, Markus Orsi, Daniel Probst, Philippe Schwaller, and Jean-Louis Reymond examine molecular relationships through more than one lens. The authors write that “Chemical space maps help visualize similarities within molecular sets.” Their contribution is to pair fingerprint-based selection with reaction-informatics analysis, so a map can show both which molecular pairs meet chosen similarity criteria and how their structures may correspond.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Drugs: From Discovery to Approval | $59.12 | Buy on Amazon |
| 2 |
|
Basic Principles of Drug Discovery and Development | $268.00 | Buy on Amazon |
| 3 |
|
Textbook of Drug Design and Discovery | $55.19 | Buy on Amazon |
| 4 |
|
Computational Drug Discovery and Design (Methods in Molecular Biology, 2714) | $139.46 | Buy on Amazon |
| 5 |
|
Drugs: From Discovery to Approval | $135.33 | Buy on Amazon |
The work demonstrates the method on three collections: FDA-approved drugs, EGFR inhibitors, and polymyxin B analogs. For the FDA-drug map, the authors report regions containing groups such as amino acids, steroids, beta-lactams, catecholamines, benzodiazepines, and prostaglandins. These groupings are patterns in a computational visualization, not claims about shared clinical effects.
Free tools Windows power users keep installed
One-click scans. No signup required.
Read the paper in Digital Discovery.
How the transformation analysis works
- Select pairs with fingerprints. The authors compare molecules using eight molecular fingerprints and select pairs that meet similarity thresholds. A pair therefore qualifies under the selected representation and cutoff; it need not be similar under every method or for every research purpose.
- Represent each pair as a difference. Each selected pair is treated as a notional transformation from one molecule to the other. A differential reaction fingerprint (DRFP) encodes the structural difference between the pair.
- Map relationships among transformations. DRFP similarities are used to arrange pairs in TMAP chemical-space visualizations, making clusters and nearest-neighbor relationships among transformations easier to inspect.
- Estimate atom correspondence. The Transformer-based RXNMapper model proposes how atoms in one molecule correspond to atoms in the other. The authors use its atom-mapping confidence distance (AMCD) as an additional signal when considering whether an implied transformation resembles a chemically feasible reaction or a more complex rearrangement.
RXNMapper’s background matters, but should not be confused with a result of the drug analysis: Orsi and colleagues describe it as trained on one million reactions documented in the USPTO dataset. That figure is the model’s reported training-corpus size, not a count of drugs or drug pairs in the study. The full-text article describes the method and model context.
#1 Best Overall
What the maps reveal—and what they do not
Similarity and transformation complexity are different questions
A fingerprint score asks how similar two molecular representations are according to a particular method. The transformation analysis asks a different question: what atom correspondence does RXNMapper propose, and how complex does the implied change appear? The authors report that RXNMapper confidence distance does not simply duplicate the fingerprint similarity measures, so it can add a distinct perspective rather than act as another name for the same score.
That distinction is useful because two molecules can satisfy a chosen similarity threshold while still implying a complicated structural rearrangement. Conversely, an atom-mapping signal is not a universal verdict on whether two drugs are “similar.” The result depends on the molecular representations, selected pairs, thresholds, and mapping model.
Hydrocodone and tetrabenazine illustrate the limits of a map
Chemistry World reports that seven of the eight fingerprints used in the study matched hydrocodone and tetrabenazine as a pair. The mapped path involved a complex double-ring formation and atom rearrangement. This is an illustrative computational mapping—not evidence of a practical synthetic route for converting one drug into the other. Chemistry World’s coverage explains the example.
Recommended Free Tools
Why this can matter in drug design
Medicinal chemists may be interested in scaffold hopping: finding compounds with similar biological activity despite different molecular frameworks. Structural maps can help surface candidate relationships for further investigation, including pairs that would not be obvious from a single comparison method. The analysis is therefore best understood as a way to organize chemical space and generate hypotheses—not as a substitute for experiments measuring activity, mechanism, safety, or clinical performance.
Rank #3
That distinction also applies to the phrase “alchemical” in the paper’s framing. It describes difficult or complex structural rearrangements in this analytical visualization; it does not mean literal transmutation, nor does it establish that a mapped transformation can be carried out in a laboratory.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to interpret a claimed drug similarity
- Ask which fingerprint and threshold were used. A statement that two molecules are similar is incomplete without the representation and selection criterion.
- Separate structure from biology. Structural resemblance alone does not establish a shared target, mechanism, indication, safety profile, or clinical effect.
- Treat atom mapping as a proposal. RXNMapper supplies a proposed atom correspondence and confidence signal, not a verified reaction route.
- Look for independent evidence of activity. Biological or therapeutic claims need evidence beyond a structural map.
The study is a cheminformatics method paper, not a clinical study or medication guide. Its findings do not support changing, substituting, or stopping a prescribed drug.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →
