A convolutional neural network for the prediction and forward design of ribozyme-based gene-control elements

Ribozyme switches are a class of RNA-encoded genetic switch that support conditional regulation of gene expression across diverse organisms. An improved elucidation of the relationships between sequence, structure, and activity can improve our capacity for de novo rational design of ribozyme switche...

Full description

Saved in:
Bibliographic Details
Published in:eLife 2021-04, Vol.10
Main Authors: Schmidt, Calvin M, Smolke, Christina D
Format: Article
Language:English
Subjects:
Aptamers
Aptamers, Nucleotide - genetics
Aptamers, Nucleotide - metabolism
Base Sequence
Binding sites
Computational and Systems Biology
Computer-Aided Design
Datasets
Design
Escherichia coli - enzymology
Escherichia coli - genetics
Gene expression
Gene Expression Regulation
Gene regulation
Genetic engineering
High-Throughput Nucleotide Sequencing
Learning algorithms
Libraries
Ligands
Machine Learning
Models, Genetic
Neural networks
Neural Networks, Computer
Nucleic Acid Conformation
Physiology
Prediction models
Regulatory sequences
RNA engineering
RNA, Catalytic - genetics
RNA, Catalytic - metabolism
Saccharomyces cerevisiae - enzymology
Saccharomyces cerevisiae - genetics
Sequence Analysis, RNA
Software
Structure-Activity Relationship
synthetic biology
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ribozyme switches are a class of RNA-encoded genetic switch that support conditional regulation of gene expression across diverse organisms. An improved elucidation of the relationships between sequence, structure, and activity can improve our capacity for de novo rational design of ribozyme switches. Here, we generated data on the activity of hundreds of thousands of ribozyme sequences. Using automated structural analysis and machine learning, we leveraged these large data sets to develop predictive models that estimate the in vivo gene-regulatory activity of a ribozyme sequence. These models supported the de novo design of ribozyme libraries with low mean basal gene-regulatory activities and new ribozyme switches that exhibit changes in gene-regulatory activity in the presence of a target ligand, producing functional switches for four out of five aptamers. Our work examines how biases in the model and the data set that affect prediction accuracy can arise and demonstrates that machine learning can be applied to RNA sequences to predict gene-regulatory activity, providing the basis for design tools for functional RNAs.
ISSN:2050-084X
2050-084X
DOI:10.7554/ELIFE.59697