Figure 1. Pie chart representation of PDB data. Sectors in the main circle represent the relative proportion of structures solved with the three main structural techniques (X‐ray, NMR, and cryo‐electron microscopy). For each technique, a pie chart is reported and divided in slices that represent the distribution of the different types of biological systems that have been structurally characterized: isolated proteins, isolated nucleic acids, and macromolecular complexes (mainly protein–protein and protein–nucleic acid complexes).

Figure 2. Bar chart of the molecular weight distribution of structures solved by NMR, X‐ray, and cryo‐electron microscopy. The total number of structures solved by each technique is equated to 100 and the frequencies of the various molecular weight classes are reported proportionally. The preferential use of NMR for low‐molecular‐weight systems and of cryo‐electron microscopy for very large assemblies clearly emerges.

Figure 3. – HSQC spectra of proteins with different degrees of folding.

1. Spectrum of a fully folded protein: all the peaks are sharp, intense, and distributed over a large chemical shift range; the number of resonances attributable to backbone amides matches the number of protein residues.
   
2. In a partially unfolded protein the chemical shift spreading of the peaks in the dimension is reduced; the most intense peaks are broad and collapsed in a narrow region centered around 8.5 ppm, typical for random‐coil structures. Peaks outside this range are weak.
   
3. The backbone amide resonances of an unfolded protein are all collapsed in a narrow range centered at about 8.3 ppm in the chemical shift dimension.
   

Peak pairs corresponding to the side‐chain NH ₂ groups of Gln and Asn are indicated by horizontal lines: their resolution also decreases with an increasing degree of unfolding.

Figure 4. Key steps that guide the iterative process of NMR structure determination. The inner cycle is generally repeated several times to correct possible errors in assignment or restraint evaluation and to enable feedback mechanisms such as the use of preliminary protein structures obtained in early cycles to supplement the initial resonance assignments and the initial collection of structural restraints. When a satisfactory number of conformers sharing good consistency with experimental data is achieved, the cycle branches off to validation/quality checks. The dashed line indicates the possible need for the further analysis of experimental data and structural restraints to improve structure quality.

Figure 5. Bundle of 30 conformers resulting from the solution structure determination of the soluble domain of a copper‐transporting protein (PDB ID : 1YJR). The bundle is graphically presented as the best superimposition of the backbone atoms of the structures forming it; side‐chains are not shown for the sake of clarity. Regular secondary structures are visualized using different colors (red for α‐helices and blue for β‐strands). In (a) and (b) each single component is represented by a separate line.

1. A smooth continuous line is used with a cartoon arrow representation of the C‐terminal part of β‐sheets to emphasize the presence of such structural elements.
   
2. A broken line connecting the C ^αs of consecutive residues is used to represent the protein backbone, thus permitting every amino acid to be picked up.
   
3. The variable precision of the bundle along the protein chain is displayed by a smooth tube with varying radius, so that local structural uncertainty is indicated by a wide tube, while a thin tube indicates the high local precision of the structure.
   

Figure 6. RMSD values and the number of distance restraints per residue for the NMR bundle of Figure 5.

1. Intraresidue, sequential, medium‐range, and long‐range distance restraints for each residue are indicated by white, light gray, dark gray, and black bars, respectively. Only the restraints used in the final calculation have been included.
   
2. The per‐residue RMSD values to the mean conformer are displayed for the backbone (filled squares) and all heavy atoms (open circles).
   

The correlation between the local number of restraints and the RMSD value is clear (e.g., at the termini or in the region around residue 15).

Figure 7. Free‐energy profile as a function of the one‐dimensional representation of the conformational space. A protein is typically characterized by a multiple‐minima free‐energy landscape. Fluctuations between the different local minima are activated processes with activation barriers of varying height that translate into a wide distribution of rate constants for the protein motions; the higher the barrier, the slower the process.

Figure 8. Typical timescale range for different types of protein motions.

Figure 9. Two‐dimensional – correlation heteronuclear multiple‐bond correlation (HMBC)‐type NMR experiment that utilizes long‐range J _C–H scalar coupling constants to correlate imino and H8 protons within a guanine base. One of the correlations in the above two‐dimensional spectrum is indicated by a straight line. The through‐bond correlations between guanine imino and H8 protons proceed via at position 5 at natural abundance using long‐range J couplings, which are indicated by green arrows.

Figure 10.

1. Multinuclear NMR study has demonstrated that the G‐quadruplex adopted by d(G ₃T ₄G ₄) exhibits two cation‐binding sites between three of its G‐quartets. The titration of tighter binding K ⁺ ions into the solution of the d(G ₃T ₄G ₄) ₂ quadruplex folded in the presence of ¹⁵NH ₄ ⁺ ions uncovered a mixed mono‐K ⁺–mono‐ ¹⁵NH ₄ ⁺ form that represents an intermediate in the conversion of di‐ ¹⁵NH ₄ ⁺ into the di‐K ⁺ form.
   
2. Riboswitches are noncoding RNA elements that directly bind small‐molecule metabolites and thereby switch gene expression on or off in response to the binding of a specific metabolite.
   

They consist of two components: an aptamer domain that interacts with a small molecule ligand and an expression platform that converts folding changes in the aptamer into altered mRNA processing.