RNAsol is a method for the prediciton of position-specific relative solvent accessible surface areas for RNAs. RNAsol was built based on improved sequence profiles from sequence-profile alignment and trained with the long short-term memory (LSTM) neural networks. Independent tests on the same datasets used by method RNAsnap show that RNAsol's accuracy is 26.5% and 136.4% higher than that of RNAsnap, for protein-bound RNAs and protein-free RNAs. The input to RNAsol is the nucleotide sequence of the RNA to be predicted. The output includes the accessible surface area for each nucleotide in the RNA sequnece. Read more about the RNAsol algorithm...