Machine learning-guided discovery and design of non-hemolytic peptides

The sequenced genomes of non-flowering land plants reveal the innovative evolutionary history of peptide signaling

An understanding of land plant evolution is a prerequisite for in-depth knowledge of plant biology. Here we extract and explore information hidden in the increasing number of sequenced plant genomes, from bryophytes to angiosperms, to elucidate a specific biological question - how peptide signaling evolved. To conquer land and cope with changing environmental conditions, plants have gone through transformations that must have required innovations in cell-to-cell communication. We discuss peptides mediating endogenous and exogenous changes by interaction with receptors activating intracellular molecular signaling. Signaling peptides were discovered in angiosperms and operate in tissues and organs such as flowers, seeds, vasculature, and 3D meristems that are not universally conserved across land plants. Nevertheless, orthologs of angiosperm peptides and receptors have been identified in non-angiosperms. These discoveries provoke questions regarding co-evolution of ligands and their receptors, and whether de novo interactions in peptide signaling pathways may have contributed to generate novel traits in land plants. The answers to such questions will have profound implications for the understanding of the evolution of cell-to-cell communication and the wealth of diversified terrestrial plants. Under this perspective we have generated, analyzed, and reviewed phylogenetic, genomic, structural, and functional data to elucidate the evolution of peptide signaling.

Prediction of Anticancer Peptides with High Efficacy and Low Toxicity by Hybrid Model Based on 3D Structure of Peptides

Recently, anticancer peptides (ACPs) have emerged as unique and promising therapeutic agents for cancer treatment compared with antibody and small molecule drugs. In addition to experimental methods of ACPs discovery, it is also necessary to develop accurate machine learning models for ACP prediction. In this study, features were extracted from the three-dimensional (3D) structure of peptides to develop the model, compared to most of the previous computational models, which are based on sequence information. In order to develop ACPs with more potency, more selectivity and less toxicity, the model for predicting ACPs, hemolytic peptides and toxic peptides were established by peptides 3D structure separately. Multiple datasets were collected according to whether the peptide sequence was chemically modified. After feature extraction and screening, diverse algorithms were used to build the model. Twelve models with excellent performance (Acc > 90%) in the ACPs mixed datasets were used to form a hybrid model to predict the candidate ACPs, and then the optimal model of hemolytic peptides (Acc = 73.68%) and toxic peptides (Acc = 85.5%) was used for safety prediction. Novel ACPs were found by using those models, and five peptides were randomly selected to determine their anticancer activity and toxic side effects in vitro experiments.