multimolecule/bprna-spot-1

bpRNA-spot是一个用于RNA二级结构预测的数据集集合，由SPOT-RNA使用。该数据集作为一个复合仓库发布，包含三个编号的组件仓库：bpRNA-spot-0（初始bpRNA分割，包括TR0、VL0和TS0）、bpRNA-spot-1（PDB迁移学习分割，包括TR1、VL1和TS1）和bpRNA-spot-2（仅NMR评估分割，包括TS2）。这些组件按顺序连接成训练集（TR0 + TR1）、验证集（VL0 + VL1）和测试集（TS0 + TS1 + TS2）。TR0/VL0/TS0分割是bpRNA-1m的子集，通过CD-HIT（CD-HIT-EST）去除序列相似性超过80%的序列，并将剩余序列按约8:1:1的比例随机分割为训练、验证和测试集。TR1/VL1/TS1分割包含用于迁移学习的高分辨率PDB RNA。TS2分割包含39个通过NMR解析的RNA，用于后训练评估。所有二级结构都以点括号表示法存储，在转换为点括号表示法之前，会移除可能导致核苷酸与多个伙伴配对的碱基对，序列文件中的非A/C/G/U符号被归一化为N。数据集模式包括id（序列标识符）、sequence（RNA序列）、secondary_structure（点括号表示法中的二级结构，可能使用超出()的括号层级表示假结）、structural_annotation（基于存储的点括号结构生成的bpRNA风格结构注释）和functional_annotation（基于存储的点括号结构生成的bpRNA风格功能注释）。bpRNA-spot is a collection of the datasets used by SPOT-RNA for RNA secondary structure prediction. The dataset is released as a composite repository, with three numbered component repositories: bpRNA-spot-0 (the initial bpRNA split, including TR0, VL0, and TS0), bpRNA-spot-1 (the PDB transfer-learning split, including TR1, VL1, and TS1), and bpRNA-spot-2 (the NMR-only evaluation split, including TS2). These components are concatenated in order to form the train set (TR0 + TR1), validation set (VL0 + VL1), and test set (TS0 + TS1 + TS2). The TR0/VL0/TS0 split is a subset of bpRNA-1m, with sequences having more than 80% sequence similarity removed using CD-HIT (CD-HIT-EST), and the remaining sequences are randomly split into training, validation, and test sets in an approximate ratio of 8:1:1. The TR1/VL1/TS1 split contains high-resolution PDB RNAs used for transfer learning. TS2 contains 39 RNAs solved by NMR and is used for post-training evaluation. All secondary structures are stored as dot-bracket notation, with base pairs that would make a nucleotide pair with multiple partners removed before conversion to dot-bracket notation, and non-A/C/G/U symbols in sequence files normalized to N. The dataset schema includes id (identifier of the sequence), sequence (RNA sequence), secondary_structure (secondary structure in dot-bracket notation, with pseudoknots potentially using bracket tiers beyond ()), structural_annotation (bpRNA-style structural annotation generated from the stored dot-bracket structure), and functional_annotation (bpRNA-style functional annotation generated from the stored dot-bracket structure).