![[2602.16643] Factorization Machine with Quadratic-Optimization Annealing for RNA Inverse Folding and Evaluation of Binary-Integer Encoding and Nucleotide Assignment](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.16643] Factorization Machine with Quadratic-Optimization Annealing for RNA Inverse Folding and Evaluation of Binary-Integer Encoding and Nucleotide Assignment
Summary
This article presents a novel method using factorization machines with quadratic-optimization annealing (FMQA) to tackle the RNA inverse folding problem, analyzing the impact of binary-integer encoding and nucleotide assignments on solution quality.
Why It Matters
The RNA inverse folding problem is critical in bioinformatics for predicting nucleotide sequences that form specific secondary structures. This study enhances existing methods by optimizing the encoding and assignment processes, potentially improving the efficiency and accuracy of RNA structure prediction, which is vital for various applications in genetics and molecular biology.
Key Takeaways
- FMQA provides a high-quality solution for RNA inverse folding with fewer evaluations.
- The choice of binary-integer encoding significantly affects the performance of RNA structure predictions.
- One-hot and domain-wall encodings outperform binary and unary encodings in terms of normalized ensemble defect value.
- Assigning guanine and cytosine to boundary integers enhances the stability of secondary structures.
- This study establishes a framework for further research in RNA folding optimization.
Computer Science > Machine Learning arXiv:2602.16643 (cs) [Submitted on 18 Feb 2026] Title:Factorization Machine with Quadratic-Optimization Annealing for RNA Inverse Folding and Evaluation of Binary-Integer Encoding and Nucleotide Assignment Authors:Shuta Kikuchi, Shu Tanaka View a PDF of the paper titled Factorization Machine with Quadratic-Optimization Annealing for RNA Inverse Folding and Evaluation of Binary-Integer Encoding and Nucleotide Assignment, by Shuta Kikuchi and 1 other authors View PDF HTML (experimental) Abstract:The RNA inverse folding problem aims to identify nucleotide sequences that preferentially adopt a given target secondary structure. While various heuristic and machine learning-based approaches have been proposed, many require a large number of sequence evaluations, which limits their applicability when experimental validation is costly. We propose a method to solve the problem using a factorization machine with quadratic-optimization annealing (FMQA). FMQA is a discrete black-box optimization method reported to obtain high-quality solutions with a limited number of evaluations. Applying FMQA to the problem requires converting nucleotides into binary variables. However, the influence of integer-to-nucleotide assignments and binary-integer encoding on the performance of FMQA has not been thoroughly investigated, even though such choices determine the structure of the surrogate model and the search landscape, and thus can directly affect solution qu...
