RNA_DESIGN

该数据集专注于 RNA逆折叠 任务，旨在通过深度学习方法探索RNA序列与结构之间的深层关联。它包含了两千余个公开的RNA骨架结构的三维原子坐标及其对应的RNA序列作为训练集，以及数百个用于评估的测试集。这些数据以numpy数组（输入坐标）和FASTA文件（输出序列）的格式提供。此数据集对于核酸适配体设计、核酶研究、RNAi药物开发、生物传感器研发和RNA纳米材料设计等领域具有重要意义，有助于开发能够从给定RNA三维结构逆向生成功能性RNA序列的算法。

This dataset focuses on the RNA inverse folding task, aiming to explore the deep correlations between RNA sequences and their structures through deep learning methods. It includes more than 2,000 publicly available 3D atomic coordinates of RNA backbone structures along with their corresponding RNA sequences as the training set, and several hundred samples as the test set for evaluation. These data are provided in the formats of NumPy arrays (for input coordinates) and FASTA files (for output sequences). This dataset holds significant implications for fields such as aptamer design, ribozyme research, RNAi drug development, biosensor development, and RNA nanomaterial design, and facilitates the development of algorithms that can reversely generate functional RNA sequences from given 3D RNA structures.