Table 3 Statistics of benchmark datasets with sequence length distribution.
From: Comprehensive datasets for RNA design, machine learning, and beyond
Benchmark | Nr of samples | Length 1–500 | \({\textrm{Length }}> {\textrm{500}}\) |
|---|---|---|---|
Eterna | 100 | 100 | 0 |
RnaBench (Inverse RNA Folding Dataset) | 68553 | 68553 | 0 |
Our dataset (loop motifs with connecting stems extracted from the RNAsolo database) | 4921 | 4840 | 81 |
Our dataset (loop motifs with connecting stems extracted from the Rfam database) | 320476 | 316832 | 3644 |
