Designing Anticancer Peptides by Constructive Machine Learning

• Published in: ChemMedChem Vol. 13; no. 13; pp. 1300 - 1302
• Main Authors: Grisoni, Francesca, Neuhaus, Claudia S., Gabernet, Gisela, Müller, Alex T., Hiss, Jan A., Schneider, Gisbert
• Format: Journal Article
• Language: English
• Published: WEINHEIM Wiley 06.07.2018; Wiley Subscription Services, Inc
• Subjects: Adenocarcinoma; Artificial intelligence; Automation; Cancer; Chemistry, Medicinal; Computer memory; de novo design; deep learning; Design; drug discovery; Erythrocytes; Learning algorithms; Life Sciences & Biomedicine; Machine learning; Memory cells; Molecular chains; Neural networks; Organic chemistry; peptide design; Peptides; Pharmacology & Pharmacy; Recurrent neural networks; Science & Technology; Selectivity; Transfer learning; peptide design; de novo design; deep learning; NK-LYSIN; drug discovery; artificial intelligence
• ISSN: 1860-7179; 1860-7187
• DOI: 10.1002/cmdc.201800204

Constructive (generative) machine learning enables the automated generation of novel chemical structures without the need for explicit molecular design rules. This study presents the experimental application of such a deep machine learning model to design membranolytic anticancer peptides (ACPs) de novo. A recurrent neural network with long short‐term memory cells was trained on α‐helical cationic amphipathic peptide sequences and then fine‐tuned with 26 known ACPs by transfer learning. This optimized model was used to generate unique and novel amino acid sequences. Twelve of the peptides were synthesized and tested for their activity on MCF7 human breast adenocarcinoma cells and selectivity against human erythrocytes. Ten of these peptides were active against cancer cells. Six of the active peptides killed MCF7 cancer cells without affecting human erythrocytes with at least threefold selectivity. These results advocate constructive machine learning for the automated design of peptides with desired biological activities. Deep learning: We report the experimental application of deep machine learning for the automated design and generation of novel membranolytic anticancer peptides. This technique, which avoids the need for explicit molecular design rules, has proven applicable to automated peptide design in a prospective setting without having to synthesize and test large sets of peptides.