At the University of Pennsylvania, a team of bioengineers has built an open-source AI platform called PeptiVerse that allows researchers to predict the therapeutic viability of peptide molecules before committing to costly laboratory synthesis. Published in Nature Communications, the tool arrives as peptide-based medicines — from GLP-1 weight-loss drugs to emerging therapeutics — have proven their real-world power, yet the path from promising molecule to working drug remains long and expensive. PeptiVerse represents a quiet but meaningful shift in how science navigates uncertainty: by moving c