 Abstract anticancer peptides, ACPs, have been identified as a promising therapeutic strategy for treating cancer. Traditional methods used to identify ACPs can be expensive and time-consuming, however. To address this issue, a team of scientists developed a novel methodology based on a computer-generated peptide library inspired by alpha-lactal Buhman, a milk protein with known anticancer properties. The team generated over 2,600 distinct peptides, each between 5 and 25 amino acids long, from alpha-lactal Buhman. These were then subjected to in-silico screening using various physical and chemical filters, as well as three different machine learning algorithms. Three peptides were identified as potential candidates, ALA, A1 and ALA, A2. In vitro testing showed that ALA, A2 was the most effective at killing lung cancer cells, while exhibiting no hemolytic side effects. Further investigation revealed that ALA, A2 induces autophagy to mediate its cytotoxicity. This study demonstrates that the sequential screening approach can be used to identify ACPs quickly and. This article was authored by Tassany Lerksithorette, Pazany Anyam, from Siri Chitfuck, and others.