Fig. 1. The direct and indirect contacts in RNA structure. The interactions of i–j, j–k, and k–m are direct contacts because they are in close distance. The interaction of i–m is indirect contact due to the transitive correlation from the tandem direct contacts. The yellow dot, red line, and blue line represent a nucleotide, direct contact, and indirect contact, respectively. The HIV-1 RNA molecule is colored in yellow with a cartoon representation (PDB code: 5L1Z. N chain).^[92]

Fig. 2. The co-evolution based RNA contact prediction. The co-evolution based contact prediction can identify the RNA intramolecular contacts from the homologous sequence across different species. The solid black lines, red dots, and green dotted lines represent RNA sequences, nucleotides, and RNA contacts, respectively. The HIV-1 RNA (PDB code: 5L1Z, N chain)^[92] is colored in yellow with a cartoon representation.

Fig. 3. The accuracy of RNA intramolecular contact prediction methods. The accuracy (positive predictive value, PPV) of MI, mfDCA, plmDCA, and DIRECT are 0.26, 0.28, 0.31, and 0.34, respectively.

Fig. 4. The accuracy of RNA intermolecular contact prediction methods. The accuracy (positive predictive value, PPV) of Rsite, Rsite2, RBind, and PRIdictor are 0.62, 0.64, 0.67, and 0.62, respectively.

Fig. 5. The structure-based RNA contact prediction. The RNA or RNA complex structural characteristic patterns can be used for the RNA contact prediction by using machine learning. The red dots, blue dots, and green dotted lines represent nucleotides, residues, and predicted RNA contacts, respectively. The HIV-1 RNA (PDB code: 5L1Z, N chain) and HIV-1 Tat protein (PDB code: 5L1Z, D chain)^[92] are colored in yellow and cyan with a cartoon representation, respectively.