Fig. 5: TS-DAR identifies transition states of a DNA motor (AlkD) translocating along a double-stranded DNA (dsDNA) over one base pair. | Nature Communications

Fig. 5: TS-DAR identifies transition states of a DNA motor (AlkD) translocating along a double-stranded DNA (dsDNA) over one base pair.

From: Exploring transition states of protein conformational changes via out-of-distribution detection in the hyperspherical latent space

Fig. 5

a Hyperspherical representations in a 3D latent space of AlkD-DNA complex system from the TS-DAR model, with dashed lines indicating the center vectors of metastable states and the rectangle box outlining the selected transition states structures. b Representative conformations of the three basins and the transition state are shown from two different point of views. AlkD and dsDNA are shown as white and gray, respectively. The \({{\rm{\alpha }}}\)-helices that are in contact with the dsDNA are shown in individual colors and the phosphor atom of the A7 phosphate group is shown as a blue sphere. The average number of hydrogen bonds between a residue of AlkD and two adjacent nucleotides of the dsDNA were calculated for each basin and transition state. c The average number of hydrogen bonds between AlkD-dsDNA that are broken (black bars) and new hydrogen bonds that are formed (red bars), during the transition from basin 1 to TS12 (left) and from TS12 to basin 2 (right). d The same as (c) except that the average number of hydrogen bonds for the transition from basin 2 to TS23 and from TS23 to basin 3 is shown. e Interaction energies between AlkD and dsDNA for basin 1, TS12, and basin 2, are shown. The interaction energies are computed by considering AlkD residues and DNA ribonucleotides that are within 3 Å of each other in at least one molecular dynamics (MD) conformation. f The same as except that AlkD-dsDNA interaction energies for basin 2, TS23, and basin 3 are shown. The average and error bars in (c)–(f) were calculated via bootstrapping of the MD simulation trajectories 100 times. For each bootstrap sample, the mean of the variable (either number of hydrogen bonds or protein-DNA interactions) was calculated for each basin and transition state. The plotted average and error bar is the average and standard deviation of all the mean values from 100 bootstrap samples, respectively. Source data are provided as a Source Data file.

Back to article page