Classification of Thyroid Peroxidase (TPO) Inhibitors Using Transfer Learning with SMILES Embeddings
Geven Piir, Sulev Sild, Eliana Spilioti, Dimitra Nikolopoulou, Effrosyni Katsanou, Ingrid Langezaal, Uko Maran · 2026
Read Classification of Thyroid Peroxidase (TPO) Inhibitors Using Transfer Learning with SMILES Embeddings by Geven Piir, Sulev Sild, Eliana Spilioti, Dimitra Nikolopoulou, Effrosyni Katsanou, Ingrid Langezaal, Uko Maran (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13273809.txt.
Description
Geven Piir and colleagues classified thyroid peroxidase inhibitors with transfer learning in a 2026 computational study. This enzyme is needed to make thyroid hormones, and blocking it is a known first step toward hormone disruption by farm chemicals. Lab testing cannot cover the huge pool of candidate chemicals.
They built models from text codes that describe molecule structure, using embeddings from a pretrained neural language model, and compared them with classic calculated descriptors. Both random forest models performed alike on training and test data, yet the embedding model proved far more sensitive on an outside validation set. Its predictions matched judgments from European safety laboratories well, though other data sources disagreed for some compounds. The authors say text based learning widens the range of chemicals such models can judge.
Catalog record
| CREATOR | Geven Piir, Sulev Sild, Eliana Spilioti, Dimitra Nikolopoulou, Effrosyni Katsanou, Ingrid Langezaal, Uko Maran |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |