Unifying pK a and Protonation Prediction with Sequence-Based Deep Learning

Charlotte Infante, Jieyu Lu, Xiaolin Pan, Song Xia, Yingkai Zhang · 2026

Read Unifying pK a and Protonation Prediction with Sequence-Based Deep Learning by Charlotte Infante, Jieyu Lu, Xiaolin Pan, Song Xia, Yingkai Zhang (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-13292215.txt.

Description

Charlotte Infante and co-authors unify acidity prediction with a language-style model in a 2026 paper. A molecule's pKa, the acidity of a specific site, steers solubility, fat liking, and binding in drug design. Trustworthy measurements at single sites are scarce, which starves training data. Terminology also differs across datasets, muddying benchmarks. The team first built a cleaned collection of nine thousand measured microscopic values with ionization notes.

Their model, T5pKa, treats molecules as text strings in a sequence-to-sequence transformer. One multitask pass lists the molecule's protonation states, gaining or losing protons as instructed. A regression step then prices the acidity between each pair of states. Across benchmarks the model matched established tools and published predictors. The selling point is unity, since one framework handles both jobs. The curated dataset is shared for others to train on.

Catalog record

CREATORCharlotte Infante, Jieyu Lu, Xiaolin Pan, Song Xia, Yingkai Zhang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13292215

Browse the full library · About the Tower of Babel