UniWave-2: A Hybrid Model for Nucleic Acid Waveform Feature Extraction Enhanced by Fourier and Wavelet Transforms
Year: 2026
Type: preprint
Abstract: Motivatio Traditional methods primarily rely on statistical features such as k-mers and GC content, making it difficult to capture complex internal relationships within sequences. Deep learning models typically rely on discrete encodings, leading to issues such as information sparsity, dimensional redundancy, and disruption of sequence continuity. Our previous UniWave-1 framework transforms sequences into waveform signals; however, it lacks comprehensive physicochemical integration and computational efficiency. Results We propose the UniWave-2 feature extraction framework. UniWave-2 incorporates three key advancements: (i) integration of hydrophobicity into the encoding scheme; (ii) enhancement of wave-form resolution through Fourier transform, coupled with wavelet transform to precisely capture both local details and global periodic patterns; and (iii) development of a lightweight multi-scale GRU model that facilitates cross-dimensional feature interactions, captures bidirectional temporal dependencies, and integrates long- and short-range patterns with spatial positional awareness, thereby yielding rich feature representations. UniWave-2 achieved competitive performance across multiple genomic tasks, while ISM analysis demonstrated that waveform encoding enables interpretable attribution without relying on gradient information. These results establish UniWave-2 as a lightweight and interpretable framework for nucleic acid sequence analysis. Availability https://github.com/qcy7226-create/Uniwave.2 Contact [email protected]
Authors: Hao Zhang, Yujun Qi, Yulan Feng, Lili Wang
Institutions: Northwest Minzu University, Northwest Normal University
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7214543262