No abstract

Secreted proteins are extracellular ligands that play key roles in paracrine and endocrine signaling, classically by binding cell surface receptors. Experimental assays to identify new extracellular ligand-receptor interactions are challenging, which has hampered the rate of novel ligand discovery. Here, using AlphaFold-multimer, we developed and applied an approach for extracellular ligand-binding prediction to a structural library of 1,108 single-pass transmembrane receptors. We demonstrate high discriminatory power and a success rate of close to 90 % for known ligand-receptor pairs where no a priori structural information is required. Importantly, the prediction was performed on de novo ligand-receptor pairs not used for AlphaFold training and validated against experimental structures. These results demonstrate proof-of-concept of a rapid and accurate computational resource to predict high-confidence cell-surface receptors for a diverse set of ligands by structural binding prediction, with potentially wide applicability for the understanding of cell-cell communication.

Text generation is the task of generating natural language, and producing outputs similar to or better than human texts. Due to deep learning’s recent success in the field of natural language processing, computer generated text has come closer to becoming indistinguishable to human writing. Genetic Algorithms have not been as popular in the field of text generation. We propose a genetic algorithm combined with text classification and clustering models which automatically grade the texts generated by the genetic algorithm. The genetic algorithm is given poorly generated texts from a Markov chain, these texts are then graded by a text classifier and a text clustering model. We then apply crossover to pairs of texts, with emphasis on those that received higher grades. Changes to the grading system and further improvements to the genetic algorithm are to be the focus of future research.

No abstract