RaptGen: A variational autoencoder with profile hidden Markov model for generative aptamer discovery

Abstract

Nucleic acid aptamers are generated by an in vitro molecular evolution method known as systematic evolution of ligands by exponential enrichment (SELEX). A variety of candidates is limited by actual sequencing data from an experiment. Here, we developed RaptGen, which is a variational autoencoder for in silico aptamer generation. RaptGen exploits a profile hidden Markov model decoder to represent motif sequences effectively. We showed that RaptGen embedded simulation sequence data into low-dimension latent space dependent on motif information. We also performed sequence embedding using two independent SELEX datasets. RaptGen successfully generated aptamers from the 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 demonstrated that RaptGen could be applied to activity-guided aptamer generation according to Bayesian optimization. We concluded that a generative method by RaptGen and latent representation are useful for aptamer discovery. Codes are available at https://github.com/hmdlab/raptgen.

DAPTEV: Deep aptamer evolutionary modelling for COVID-19 drug design

“…This research was continued by Park et al and a similar conclusion was drawn for RNA aptamers specifically [39]. In [40], Iwano et al propose RaptGen which is similar to [37] and [39]. They too utilized DeepBind, however, RaptGen implemented a CNN-LSTM as their encoder and a profile hidden Markov model as their decoder.…”

“…After obtaining 849 unique known aptamer sequences [13,51,52], only 390 met the condition of being between the minimum (20) and maximum (40)…”

“…After obtaining 849 unique known aptamer sequences [13,51,52], only 390 met the condition of being between the minimum (20) and maximum (40) sequence lengths. Of those 390 aptamer, 344 contained at least one connection (base pairing) in their secondary structures.…”

Representation learning applications in biological sequence analysis

“…Embedding biological sequences using NLP overcomes these difficulties and outperforms existing methods in many tasks, such as function, structure, localization, and disorder prediction (Table 1). In addition to these general biological tasks, representation learning has also been used to solve specific problems, such as RNA aptamer optimization [98], viral mutation prediction [56], and venom toxin prediction [36]. In these studies, representation learning of biological sequences could capture biophysical and biochemical properties of biological systems, and representation learning may reveal the grammar of life.…”

DAPTEV: Deep aptamer evolutionary modelling for COVID-19 drug design

“…and a similar conclusion was drawn for RNA aptamers specifically [ 39 ]. In [ 40 ], Iwano et al . propose RaptGen which is similar to [ 37 ] and [ 39 ].…”