Concetta Press

@article{ciaccio2026TIF,
  title={Glass Box Machine Learning to Aid the Design of Small Peptide Agonists from Very Sparse Data: The Example of Erythropoietin Analogues},
  author={Robson, Barry},
  journal={Journal of Biomedical Informatics and AI},
  volume={1},
  number={1},
  pages={4},
  year={2026},
  doi={10.5281/zenodo.20073063},
  publisher={Concetta Press},
  url={https://concettapress.net/article_glass_box_machine_learning.html},
  pdf={https://concettapress.net/pdf/JBIAI_1_1_4(2026).pdf}
}

Abstract. This paper presents a perspective on interpretable approaches to peptide design when, as is often the case, initial data is very sparse. It provides a worked example with discussion for the very earliest stage that one may can consider a “hunch phase”, when data is sparse and often circumstantial, and suspected activity and inactivity for many peptides is more correctly described as unknown. The very earliest steps involve bioinformatics and related aspects of computational chemistry approaches that are essentially standard, but their purpose is to generate all possible information for a final explanatory Glass Box AI approach involving use of a Theory of Expected Information developed for sparse data which is much less commonly described. They still have some novel or unusual features which make them well suited to that task, and glass box AI applied to peptides can itself be considered a recent form of bioinformatics. The present discussion can also be considered as an early step in the design of traditional therapeutics, i.e. small organic “in-a-pill” drugs. This is because biologically active peptides can provide clues for design of small organics, help establish laboratory assays, and provide important information as to the action of agonists and antagonists, and as to safety. Also providing early data are cases where peptides have reached the marketing stage, but still have disadvantages, even being withdrawn, so increasing the demand for more traditional therapeutics that may not have the same problems. In other cases, there is data from viable peptide therapeutics where use is confined to one or few countries. Erythropoietic peptides for treatment of anemia provide an example of all these cases.

Keywords: peptides; peptidomimetics; artificial intelligence; anemia; erythropoietin; sparse data