The dataset for the training, validation and testing of the RNARL framework

This dataset consists of a total of 21 CSV files. Among them, 18 files correspond to the protein and associated mRNA CDS datasets for six organisms: Chlamydomonas reinhardtii, Thermococcus kodakarensis KOD1, Homo sapiens, Arabidopsis thaliana, Escherichia coli, and Penicillium chrysogenum. Each organism has a dedicated training set, validation set, and test set, resulting in 18 files (6 organisms × 3 sets each). The remaining three files contain human circular RNA CDS sequences and their corresponding protein datasets, also divided into a training set, validation set, and test set.

In each CSV file, one protein sequence is paired with two CDS sequences: one generated by LinearDesign with a lambda parameter of 0, and the other with a lambda parameter of 1. The files include the following fields: protein sequence, CDS sequence, protein sequence length, CDS sequence length, CDS secondary structure, CDS minimum free energy (MFE), and CDS codon adaptation index (CAI).