Machine Learning Reveals Structural Properties of Monosaccharides and their Peptide Conjugates
Year: 2024
Type: conference-paper
Abstract: European Peptide Society is kindly acknowledged for the awarded registration grant to Mr. Michele Casoria These 180 features were collected into a matrix describing all the sequons, UMAP reduction to three dimensions was applied, followed by clustering of the structure using the DBSCAN algorithm, with an epsilon value of 0.4 and a minimum sample size of 4. The resulting evaluation metrics are as follows: Silhouette Score: 0.69, Cluster Compactness: 0.04, and Cluster Separation: 12.35.Additionally, we identified 2625 clusters and 21 structures in cluster -1 as noise.
Source: European Peptide Symposium (EPS)
Institution: University of Florence
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W4402745231