License: 
Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.WABI.2023.21
URN: urn:nbn:de:0030-drops-186479
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2023/18647/
Gu, Xinyu ;
Qi, Yuanyuan ;
El-Kebir, Mohammed
Balancing Minimum Free Energy and Codon Adaptation Index for Pareto Optimal RNA Design
Abstract
The problem of designing an RNA sequence v that encodes for a given target protein w plays an important role in messenger RNA (mRNA) vaccine design. Due to codon degeneracy, there exist exponentially many RNA sequences for a single target protein. These candidate RNA sequences may adopt different secondary structure conformations with varying minimum free energy (MFE), affecting their thermodynamic stability and consequently mRNA half-life. In addition, species-specific codon usage bias, as measured by the codon adaptation index (CAI), also plays an essential role in translation efficiency. While previous works have focused on optimizing either MFE or CAI, more recent works have shown the merits of optimizing both objectives. Importantly, there is a trade-off between MFE and CAI, i.e. optimizing one objective is at the expense of the other. Here, we formulate the Pareto Optimal RNA Design problem, seeking the set of Pareto optimal solutions for which no other solution exists that is better in terms of both MFE and CAI. We introduce DERNA (DEsign RNA), which uses the weighted sum method to enumerate the Pareto front by optimizing convex combinations of both objectives. DERNA uses dynamic programming to solve each convex combination in O(|w|³) time and O(|w|²) space. Compared to a previous approach that only optimizes MFE, we show on a benchmark dataset that DERNA obtains solutions with identical MFE but superior CAI. Additionally, we show that DERNA matches the performance in terms of solution quality of LinearDesign, a recent approach that similarly seeks to balance MFE and CAI. Finally, we demonstrate our method’s potential for mRNA vaccine design using SARS-CoV-2 spike as the target protein.
BibTeX - Entry
@InProceedings{gu_et_al:LIPIcs.WABI.2023.21,
author = {Gu, Xinyu and Qi, Yuanyuan and El-Kebir, Mohammed},
title = {{Balancing Minimum Free Energy and Codon Adaptation Index for Pareto Optimal RNA Design}},
booktitle = {23rd International Workshop on Algorithms in Bioinformatics (WABI 2023)},
pages = {21:1--21:20},
series = {Leibniz International Proceedings in Informatics (LIPIcs)},
ISBN = {978-3-95977-294-5},
ISSN = {1868-8969},
year = {2023},
volume = {273},
editor = {Belazzougui, Djamal and Ouangraoua, A\"{i}da},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/opus/volltexte/2023/18647},
URN = {urn:nbn:de:0030-drops-186479},
doi = {10.4230/LIPIcs.WABI.2023.21},
annote = {Keywords: Multi-objective optimization, dynamic programming, RNA sequence design, reverse translation, mRNA vaccine design}
}
