.jpeg)
Structural bioinformatics is a branch of bioinformatics that uses computers and computational tools to study the three-dimensional (3D) structures of proteins, DNA, RNA, and other biological molecules. The structure of a protein helps determine its function. Structural bioinformatics allows researchers to visualize molecular structures, analyze their properties, check their quality, and compare them with other structures.
For example, scientists can study a protein's 3D structure to understand how it interacts with other molecules or how it performs its biological function. If you want to start from the sequence level, see our guide on protein sequence analysis.
Importance of Structural Bioinformatics
Structural bioinformatics helps researchers understand the relationship between the structure and function of biological molecules.
Its main applications include:
- Protein structure analysis: Studying the 3D shape of proteins. (Sequence-based predictions of properties such as pI, signal peptides, and secondary structure are covered in Protein Sequence Analysis.)
- Drug discovery: Understanding how potential drugs interact with target proteins.
- Disease research: Investigating structural changes associated with diseases.
- Protein classification: Grouping proteins based on their structural similarities. (For similarity at the sequence level, see BLAST and Homology Search.)
- Molecular visualization: Displaying and examining biological structures using computer programs.
What Is the Protein Data Bank (PDB)?
The Protein Data Bank (PDB) is a public database that stores the 3D structures of proteins, DNA, RNA, and their complexes. Researchers can access these structures to study molecular shapes and biological functions. To see how the PDB fits among other resources, read Biological Databases. Each PDB entry has a unique identifier, such as 1UBQ, which represents a particular structure.
Researchers can download structure files from the RCSB Protein Data Bank for visualization and analysis. For a broader look at finding and downloading datasets, see Accessing Public Bioinformatics Data and Sequence Retrieval and Basic Analysis.
Methods Used to Determine Protein Structures
Scientists use different experimental methods to determine the 3D structures of biological molecules.
| Method | Definition |
|---|---|
| X-ray crystallography | Uses X-ray diffraction patterns from crystals to determine molecular structures. |
| NMR spectroscopy | Uses signals from atomic nuclei to study molecular structures and their movements in solution. |
| Cryo-electron microscopy (Cryo-EM) | Uses images of frozen biological samples to reconstruct 3D structures. |
What Is Resolution in Protein Structure?
Resolution describes the level of detail visible in an experimentally determined molecular structure. It is commonly measured in angstroms (Å). In general, a lower resolution value indicates greater structural detail.
For example, a structure with a resolution of 1.8 Å generally provides more detail than one with a resolution of 3.0 Å. However, resolution alone does not determine the overall quality of a structure. Resolution is also one of the checks used in protein structure validation.
What Is a PDB File?
A PDB file is a text-based file format used to store information about the atoms and structure of biological molecules. It contains details such as atom names, amino acid residues, chain identifiers, and atomic coordinates. It is one of several formats covered in Bioinformatics File Formats. Researchers can open a PDB file in a text editor or use molecular visualization software such as PyMOL to examine the structure.
To read and process PDB files with code, see Python for Bioinformatics and Python, Biopython and Data Libraries in Bioinformatics. To handle many structure files from the terminal, see Linux Commands for Bioinformatics.
Important Records in a PDB File
A PDB file contains different record types, each providing specific information.
| Record | Definition |
|---|---|
| HEADER | Contains general information about the structure. |
| TITLE | Describes the structure. |
| COMPND | Provides information about the molecular components. |
| SOURCE | Identifies the biological source of the molecule. |
| EXPDTA | Specifies the experimental method used to determine the structure. |
| REMARK | Contains additional information, including experimental and validation details. |
| SEQRES | Lists the amino acid sequence of a protein chain (see Sequence Retrieval and Basic Analysis). |
| HELIX | Identifies alpha-helical regions (see secondary structure prediction). |
| SHEET | Identifies beta-sheet regions. |
| ATOM | Stores coordinates and related information for standard atoms in the structure. |
| HETATM | Stores information about non-standard residues and other atoms, such as water molecules, ions, and ligands. |
| TER | Marks the end of a chain. |
| END | Marks the end of the file. |
What Are ATOM and HETATM Records?
ATOM records contain information about standard atoms in a molecular structure. They include atomic coordinates that describe the position of each atom in three-dimensional space.
HETATM records contain information about atoms belonging to non-standard residues and other molecular components, such as water molecules, metal ions, and ligands.
A typical atom record includes the following information:
- Atom serial number
- Atom name
- Residue name
- Chain identifier
- Residue number
- X, Y, and Z coordinates
- Occupancy
- B-factor
- Element name
What Are Occupancy and B-Factor?
Occupancy indicates the proportion of a modeled atom's population assigned to a particular position. A value of 1.00 generally represents full occupancy in the structural model.
B-factor, also called the temperature factor, describes the positional disorder of an atom in an experimentally determined structure. Higher values can indicate greater movement or uncertainty in its position, although the interpretation depends on the structure and experimental data.
What Is the mmCIF File Format?
mmCIF is a file format used to store detailed information about biological structures. It supports more extensive data than the older PDB text format and is widely used for structures available through the Protein Data Bank (see What Is the Protein Data Bank?).
Many molecular visualization programs, including PyMOL, can work with both PDB and mmCIF files. Other common formats such as FASTA and FASTQ are explained in Bioinformatics File Formats.
What Is PyMOL?
PyMOL is a molecular visualization program used to view, rotate, analyze, and create images of biological molecules. Researchers use PyMOL to examine protein structures, identify secondary structures, highlight specific residues, and measure distances between atoms.
How to Visualize a Protein Using PyMOL
The following commands demonstrate how to load and display a protein structure.
Step 1: Load a protein structure
fetch 1ubq
This command downloads the ubiquitin structure identified by PDB ID 1UBQ from the Protein Data Bank.
Step 2: Display the structure as a cartoon
hide everything show cartoon bg_color white
The cartoon representation makes it easier to identify helices, beta sheets, and loops.
Step 3: Change the colors of secondary structures
color red, ss h color yellow, ss s color green, ss l+
These commands color helices red, beta strands yellow, and loops green.
Step 4: Highlight a specific residue
show sticks, resi 44 zoom resi 44
These commands display residue 44 in a stick representation and zoom in on it.
Step 5: Save the structure image
png ubiquitin.png, width=1200, height=900, dpi=300, ray=1
This command saves an image of the molecular structure.
Common PyMOL Commands
| Command | Purpose |
|---|---|
| fetch | Downloads a structure from the PDB. |
| load | Loads a local structure file. |
| show | Displays a selected representation. |
| hide | Hides a representation. |
| color | Changes the color of a structure or selection. |
| select | Creates a named selection of atoms. |
| label | Adds labels to selected atoms or residues. |
| distance | Measures the distance between selected atoms. |
| zoom | Zooms in on a structure or selection. |
| orient | Adjusts the view to orient the selected structure. |
| png | Saves an image of the displayed structure. |
| save | Saves a structure or PyMOL session, depending on the command and file format. |
What Is a Ramachandran Plot?
A Ramachandran plot is a graphical representation of the backbone dihedral angles of amino acid residues in a protein. It helps researchers examine whether the backbone conformations of a protein are structurally reasonable. (New to amino acids and peptide bonds? See Biology Basics for Coders.)
The plot uses two angles:
- Phi (φ): Rotation around the N–CA bond.
- Psi (ψ): Rotation around the CA–C bond.
These angles determine the possible conformations of the protein backbone.
Main Regions of a Ramachandran Plot
| Region | Description |
|---|---|
| Alpha-helix region | Contains angle combinations commonly associated with alpha helices. |
| Beta-sheet region | Contains angle combinations commonly associated with beta strands and sheets. |
| Left-handed helix region | Contains conformations associated with left-handed helices. |
| Disallowed regions | Contain conformations that are generally unfavorable because of steric clashes or other structural constraints. |
Importance of a Ramachandran Plot
A Ramachandran plot helps researchers evaluate protein structure quality by identifying residues with unusual backbone angles.
Most residues in a well-modeled protein are expected to fall within favored or allowed regions. Residues outside these regions are called outliers and may require further investigation.
However, an outlier does not automatically mean that the structure is incorrect. Its position should be evaluated using the experimental evidence and other validation results.
Role of Glycine and Proline
Glycine has a small side chain consisting of a hydrogen atom. This gives it greater conformational flexibility than most other amino acids.
Proline has a ring-shaped side chain connected to its backbone nitrogen. This restricts its backbone flexibility and limits the range of conformations it can adopt.
These differences affect the positions of glycine and proline residues on a Ramachandran plot.
What Is Protein Structure Validation?
Protein structure validation is the process of checking whether a protein structure is consistent with experimental data and expected chemical and structural properties.
A structure may contain errors in atomic positions, bond geometry, side-chain conformations, or backbone angles. Validation helps researchers identify such problems before using the structure in further analysis. Checks such as the Ramachandran plot and resolution are part of this process.
Important Protein Structure Validation Checks
| Check | Purpose |
|---|---|
| Resolution | Evaluates the level of detail in an experimentally determined structure. |
| R-work and R-free | Assess how well a crystallographic model agrees with experimental diffraction data. |
| Ramachandran outliers | Identify unusual backbone angle combinations (see Ramachandran plot). |
| Rotamer outliers | Identify unusual side-chain conformations. |
| Clashscore | Measures serious overlaps between atoms. |
| Bond lengths and angles | Check whether molecular geometry is reasonable. |
| Electron-density fit | Evaluates how well the model is supported by the experimental electron-density map. |
| Missing residues | Identifies parts of the protein that are not represented in the structural model. |
What Are R-Work and R-Free?
R-work and R-free are measures used to evaluate models determined using X-ray crystallography (see methods used to determine structures).
R-work measures the agreement between the model and the diffraction data used during model refinement.
R-free measures agreement with a separate set of diffraction data that was excluded from refinement. It helps researchers assess whether the model generalizes beyond the data used to build it.
Both values should be interpreted together with the resolution, model complexity, and other validation results. For the statistics background behind such measures, see Biostatistics for Bioinformatics.