Generative models for antimicrobial peptide design: auto-encoders and beyond

Lukas Beierle, Julian Hahnfeld, Alexander Goesmann, Reihaneh Mostolizadeh, Franz Cemič · 2026

Read Generative models for antimicrobial peptide design: auto-encoders and beyond by Lukas Beierle, Julian Hahnfeld, Alexander Goesmann, Reihaneh Mostolizadeh, Franz Cemič (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-13181977.txt.

Description

Lukas Beierle, Julian Hahnfeld, Alexander Goesmann, Reihaneh Mostolizadeh, and Franz Cemič present a study dated 2026 on using AI to design antimicrobial peptides. These short protein chains can kill germs and tend to provoke less drug resistance than standard antibiotics. With resistant germs spreading, fast ways to invent new candidates are in demand.

The team compared four generative model types, including two auto-encoders, a recurrent network, and a language model. Each was asked to create new peptide sequences. All produced chains with traits close to natural ones. The auto-encoders did best overall, and one variant gave the most varied and balanced output. Checks on the learned space showed the models were not just copying their training data.

Catalog record

CREATORLukas Beierle, Julian Hahnfeld, Alexander Goesmann, Reihaneh Mostolizadeh, Franz Cemič
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE74 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13181977

Browse the full library · About the Tower of Babel