AggBERT: Best in Class Prediction of Hexapeptide Amyloidogenesis with a Semi-Supervised ProtBERT Model

The prediction of peptide amyloidogenesis is a challengingproblemin the field of protein folding. Large language models, such as theProtBERT model, have recently emerged as powerful tools in analyzingprotein sequences for applications, such as predicting protein structureand function. In this article, we describe the use of a semisupervisedand fine-tuned ProtBERT model to predict peptide amyloidogenesis fromsequences alone. Our approach, which we call AggBERT, achieved state-of-the-artperformance, demonstrating the potential for large language modelsto improve the accuracy and speed of amyloid fibril prediction oversimple heuristics or structure-based approaches. This work highlightsthe transformative potential of machine learning and large languagemodels in the fields of chemical biology and biomedicine.

Automated summary

This article describes the use of the AggBERT approach, utilizing the ProtBERT model, to predict peptide amyloidogenesis. The results demonstrate that large language models have the potential to improve the accuracy and speed of amyloid fibril prediction in chemical biology and biomedicine.