Researchers from the Graduate School of Advanced Science and Engineering at Waseda University, Japan claim to have introduced RaptGen, a variational autoencoder (VAE) that can be used for the generation of aptamers. VAE, a type of machine learning approach, has been shown to be beneficial in the discovery of other small molecules.
The scientists published their paper “Generative discovery of aptamers using RaptGen” in Computational science of nature and explain how RaptGen uses a VAE with a hidden Markov model decoder to create latent spaces in which sequences can form clusters.
Using this latent representation, RaptGen was able to generate aptamers that were not even included in the original sequencing data or the HT-SELEX dataset.
Aptamers are a type of oligonucleotide that can selectively bind to specific targets such as proteins, peptides, carbohydrates, viruses, toxins, metal ions, and living cells. Since they are similar to antibodies, they have a variety of uses in the fields of biosensors, therapeutics and diagnostics. Since they are similar to antibodies, they have a variety of uses in the fields of biosensors, therapeutics and diagnostics. However, unlike antibodies, aptamers do not induce an immune reaction in our body and are easy to synthesize and modify. Additionally, the three-dimensional folding structure of an aptamer allows it to bind to a wider range of targets.
“Nucleic acid aptamers are generated by an in vitro molecular evolution method known as systematic evolution of ligands by exponential enrichment (SELEX). Various candidates are limited by the actual sequencing data of an experiment. Here, we developed RaptGen, which is a variational autoencoder for in silico aptamer generation,” the researchers write.
“RaptGen leverages a hidden Markov model decoder to efficiently represent pattern sequences. We showed that RaptGen embeds simulation sequence data into a low-dimensional latent space based on pattern information. We also performed sequence embedding using two independent SELEX datasets. RaptGen successfully generated aptamers from latent space even though they were not included in high-throughput sequencing. RaptGen could also generate a truncated aptamer with a short learning model.
“We have demonstrated that RaptGen can be applied to activity-guided aptamer generation using Bayesian optimization. We concluded that a generative method by RaptGen and a latent representation are useful for the discovery of aptamers.
“RaptGen first visualizes a latent space with a sequence pattern, then generates several new aptamer sequences via this latent space,” says Michiaki Hamada, PhD, professor, in describing how RaptGen can boost the discovery of aptamers.
“For example, it searches for optimized aptamer sequences in latent space by considering additional information after analyzing the activity of a subset of sequences. Additionally, RaptGen allows the design of shortened (or truncated) aptamer sequences. »
The team also evaluated RaptGen’s performance using real-world data, subjecting it to data from two independent HT-SELEX datasets. RaptGen could generate aptamer derivatives in an activity-driven manner and provide opportunities to optimize their business.
“This is important because it means RaptGen can generate sequences with desired properties, such as inhibiting certain enzymes or protein-protein interactions,” adds Hamada. “The application of these molecules could open many doors in the future.”
The scientists plan to conduct extensive studies to assess whether alternative models can improve the performance of RaptGen and whether RaptGen could advance the generation of RNA aptamers using RNA sequences. The only drawbacks of using RaptGen are high computational cost and increased training time, both of which can be improved in further studies, according to Hamada.
“To our knowledge, RaptGen is the only data-driven method that can design and optimize truncated aptamers directly from HT-SELEX data,” Hamada notes. “We believe that in due course, RaptGen will be recognized as a key tool for effective aptamer discovery.”
#computer #model #developed #generate #aptamers