Finding ways to more comprehensively explore the sequence space of complex functional motifs is an important and unresolved question in nucleic acid engineering. Standard approaches use libraries in which a single variant of a motif is randomly mutagenized at a low level. This provides comprehensive coverage of sequence space over short mutational distances, but only limited information about more distant variants. Here we describe a new approach that uses libraries made up of sequences consistent with the multiple constraints of a desired target motif. Functional variants are rapidly identified in a single round of selection followed by high-throughput sequencing, and rules relating sequence to function are elucidated using machine learning. This method was tested using a fluorescent deoxyribozyme recently discovered in our group called Aurora. Single-step selections showed that a secondary structure library based on Aurora contained ~40-fold more unique catalytic sequences than one generated by random mutagenesis. Furthermore, models developed by machine learning could quantitatively predict read numbers and identify the most active variants using small subsets of sequences as training sets. By combining secondary structure libraries, selection, and machine learning in this way, sequence space can be explored far more quickly and efficiently than in standard approaches.

Czech Institute of Informatics Robotics and Cybernetics Czech Technical University Prague 160 00 Czech Republic

Department of Genetics and Microbiology Faculty of Science Charles University Prague Prague 128 44 Czech Republic

Department of Informatics and Chemistry University of Chemistry and Technology Prague 166 28 Czech Republic

Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences Prague 166 10 Czech Republic

Zobrazit více v PubMed

Tuerk  C, Gold  L. Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science. 1990;249:505–10. 10.1126/science.2200121. PubMed DOI

Robertson  DL, Joyce  GF. Selection PubMed DOI

Ellington  AD, Szostak  JW. PubMed DOI

Silverman  SK. Catalytic DNA: scope, applications, and biochemistry of deoxyribozymes. Trends Biochem Sci. 2016;41:595–609. 10.1016/j.tibs.2016.04.010. PubMed DOI PMC

Micura  R, Höbartner  C. Fundamental studies of functional nucleic acids: aptamers, riboswitches, ribozymes and DNAzymes. Chem Soc Rev. 2020;49:7331–53. 10.1039/D0CS00617C. PubMed DOI

Zhou  J, Rossi  J. Aptamers as targeted therapeutics: current potential and challenges. Nat Rev Drug Discov. 2017;16:181–202. 10.1038/nrd.2016.199. PubMed DOI PMC

Curtis  EA. Pushing the limits of nucleic acid function. Chemistry A European J. 2022;28:e202201737. 10.1002/chem.202201737. PubMed DOI PMC

Ekland  EH, Bartel  DP. The secondary structure and sequence optimization of an RNA ligase ribozyme. Nucleic Acids Res. 1995;23:3231–8. 10.1093/nar/23.16.3231. PubMed DOI PMC

Knight  R, Yarus  M. Analyzing partially randomized nucleic acid pools: straight dope on doping. Nucleic Acids Res. 2003;31:e30. 10.1093/nar/gng030. PubMed DOI PMC

Bashir  A, Yang  Q, Wang  J  et al.  Machine learning guided aptamer refinement and discovery. Nat Commun. 2021;12:2366. 10.1038/s41467-021-22555-9. PubMed DOI PMC

Beck  JD, Roberts  JM, Kitzhaber  JM  et al.  Predicting higher-order mutational effects in an RNA enzyme by machine learning of high-throughput experimental data. Front Mol Biosci. 2022;9:893864. 10.3389/fmolb.2022.893864. PubMed DOI PMC

Rotrattanadumrong  R, Yokobayashi  Y. Experimental exploration of a ribozyme neutral network using evolutionary algorithm and deep learning. Nat Commun. 2022;13:4847. 10.1038/s41467-022-32538-z. PubMed DOI PMC

Volek  M, Kurfürst  J, Drexler  M  et al.  Aurora: a fluorescent deoxyribozyme for high-throughput screening. Nucleic Acids Res. 2024;52:9049–61. 10.1093/nar/gkae467. PubMed DOI PMC

Sgallová  R, Curtis  EA. Secondary structure libraries for artificial evolution experiments. Molecules. 2021;26:1671. 10.3390/molecules26061671. PubMed DOI PMC

Streckerová  T, Kurfürst  J, Curtis  EA. Single-round deoxyribozyme discovery. Nucleic Acids Res. 2021;49:6971–81. 10.1093/nar/gkab504. PubMed DOI PMC

Svehlova  K, Lukšan  O, Jakubec  M  et al.  Supernova: a deoxyribozyme that catalyzes a chemiluminescent reaction. Angew Chem Int Ed. 2022;61:e202109347. 10.1002/anie.202109347. PubMed DOI PMC

Jakubec  M, Pšenáková  K, Svehlova  K  et al.  Optimizing the chemiluminescence of a light-producing deoxyribozyme. ChemBioChem. 2022;23:e202200026. 10.1002/cbic.202200026. PubMed DOI

Volek  M, Kurfürst  J, Kožíšek  M  et al.  Apollon: a deoxyribozyme that generates a yellow product. Nucleic Acids Res. 2024;52:9062–75. 10.1093/nar/gkae490. PubMed DOI PMC

Gutell  RR, Power  A, Hertz  GZ  et al.  Identifying constraints on the higher-order structure of RNA: continued development and application of comparative sequence analysis methods. Nucleic Acids Res. 1992;20:5785–95. 10.1093/nar/20.21.5785. PubMed DOI PMC

Gutell  RR. Ten lessons with Carl Woese about RNA and comparative analysis. RNA Biol. 2014;11:254–72. 10.4161/rna.28718. PubMed DOI

Curtis  EA, Bartel  DP. Synthetic shuffling and PubMed DOI PMC

Akoopie  A, Müller  UF. The NTP binding site of the polymerase ribozyme. Nucleic Acids Res. 2018;46:10589–97. PubMed PMC

Kaufman  S. The Origin of Order.  Oxford, UK:  Oxford University Press;  1993.

Chiu  DK, Kolodziejczak  T. Inferring consensus structure from nucleic acid sequences. Comput Appl Biosci. 1991;7:347–52. PubMed

Pitt  JN, Ferré-D’Amaré  AR. Rapid construction of empirical RNA fitness landscapes. Science. 2010;330:376–9. 10.1126/science.1192001. PubMed DOI PMC

Jiménez  JI, Xulvi-Brunet  R, Campbell  GW  et al.  Comprehensive experimental fitness landscape and evolutionary network for small RNA. Proc Natl Acad Sci USA. 2013;110:14984–9. PubMed PMC

Domingo  J, Diss  G, Lehner  B. Pairwise and higher-order genetic interactions during the evolution of a tRNA. Nature. 2018;558:117–21. 10.1038/s41586-018-0170-7. PubMed DOI PMC

Pressman  AD, Liu  Z, Janzen  E  et al.  Mapping a systematic ribozyme fitness landscape reveals a frustrated evolutionary network for self-aminoacylating RNA. J Am Chem Soc. 2019;141:6213–23. 10.1021/jacs.8b13298. PubMed DOI PMC

Jumper  J, Evans  R, Pritzel  A  et al.  Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596:583–9. 10.1038/s41586-021-03819-2. PubMed DOI PMC

Angermueller  C, Pärnamaa  T, Parts  L  et al.  Deep learning for computational biology. Mol Syst Biol. 2016;12:878. 10.15252/msb.20156651 . PubMed DOI PMC

Auslander  N, Gussow  AB, Koonin  EV. Incorporating machine learning into established bioinformatics frameworks. Int J Mol Sci. 2021;22:2903. 10.3390/ijms22062903. PubMed DOI PMC