Enriched Human Macromolecule Structure Dataset

Columns

• PDB ID: Unique identifier for each protein structure.
• Experimental Method: The technique used to determine the structure, primarily X-ray diffraction.
• Matthews Coefficient: An estimate of the crystal's volume occupied by protein versus solvent.
• Percent Solvent Content: Percentage of solvent within the crystal.
• Crystallization Method: The technique used for crystal growth, such as vapor diffusion.
• pH: The pH level at which crystallization occurred.
• Crystal Growth Procedure: A detailed description of the crystal growth process.
• Temp (K): Crystallization temperature in Kelvin.
• Deposition Date: The date the structure was submitted to the PDB.
• Release Date: The date the structure was made public.
• Number of Non-Hydrogen Atoms per Deposited Model: Count of non-hydrogen atoms in the structure.
• Total Number of Polymer Instances (Chains): The number of unique polymer chains.
• Total Number of Polymer Residues per Deposited Model: Total amino acid residues in the model.
• Number of Water Molecules per Deposited Model: Total water molecules in the structure.
• Disulfide Bond Count per Deposited Model: The number of disulfide bonds.
• Molecular Weight per Deposited Model: Molecular weight of the entire structure.
• Number of Distinct Protein Entities: The count of unique protein entities.
• Refinement Resolution (Å): Resolution of the X-ray experiment in Ångströms.
• Structure Determination Methodology: The overall method used, e.g., 'experimental'.
• Average B Factor: A measure of atomic displacement, indicating flexibility.
• R Free: A validation metric for the crystallographic model quality.
• R Work: A measure of how well the model fits experimental data.
• Structure Title: The title assigned by the researchers.
• Sequence: The amino acid sequence of the protein.
• Entity Polymer Type: The type of polymer, e.g., 'Protein'.
• Polymer Entity Sequence Length: The length of the polymer sequence.
• Entity Macromolecule Type: The type of macromolecule, e.g., 'polypeptide(L)'.
• Total Number of Polymer Entity Instances (Chains) per Entity: Number of chains per polymer entity.
• Molecular Weight (Entity): Molecular weight of an individual polymer entity.
• Macromolecule Name: The name of the macromolecule.
• EC Number: The Enzyme Commission number for enzyme classification.
• EC Provenance Source: The source of the enzyme classification.
• Source Organism: The organism from which the protein was derived.
• Taxonomy ID: Identifier for the source organism's taxonomy.
• Total Number of Polymer Residues per Assembly: Total residues in the full assembly.
• Total Number of Polymer Instances (Chains) per Assembly: Total chains in the full assembly.
• Oligomeric Count: The number of subunits in the oligomeric state.
• Assembly ID: A unique identifier for the assembly.
• Oligomeric State: The functional form of the protein (e.g., 'Monomer').
• Stoichiometry: The ratio of components in the protein assembly.
• Ligand ID: Identifier for any bound ligand.
• Ligand Formula: The chemical formula of the ligand.
• Ligand MW: The molecular weight of the ligand.
• Ligand Name: The common name of the ligand.
• InChI: The International Chemical Identifier for the ligand.
• Ligand of Interest: Indicates if a ligand is of special interest.
• Number of Residues: Total count of amino acid residues (newly added).
• Number of Chains: Count of distinct chains in the structure (newly added).
• Helix Count: Number of alpha-helices (newly added).
• Sheet Count: Number of beta-sheets (newly added).
• Coil Count: Number of random coil regions (newly added).

Distribution

• Format: A single CSV file named RCSB_PDB_Macromolecular_Structure_Dataset.csv.
• Size: 11.36 MB.
• Structure: The dataset is tabular and contains 11,832 records (rows) and 46 columns.

Usage

• Protein-ligand binding prediction.
• Oligomeric state analysis and prediction.
• Enzyme classification using EC numbers.
• Protein secondary structure prediction.
• Assessment of protein structure quality.
• Protein stability and protein-protein interaction prediction.
• Protein domain analysis and evolutionary analysis.
• Prediction of optimal protein crystallization conditions.

Coverage

• Geographic: Not applicable, as the data is molecular. The source organism is predominantly Homo sapiens (human).
• Time Range: The dataset includes protein structures with release dates from January 2015 to September 2023.
• Demographic: The data pertains to human (Homo sapiens) proteins.

License

Who Can Use It

• Bioinformaticians and Computational Biologists: For research in structural biology, protein function prediction, and evolutionary analysis.
• Data Scientists and Machine Learning Engineers: For developing and training models on tasks like enzyme classification, ligand binding prediction, and protein stability analysis.
• Pharmaceutical Researchers: For drug discovery, identifying potential binding sites, and analysing protein-ligand interactions.
• Academics and Students: For educational purposes and research projects in biochemistry, molecular biology, and data science.

Dataset Name Suggestions

• Human Protein Structures for Machine Learning
• RCSB PDB Human Protein Structural Features
• Enriched Human Macromolecule Structure Dataset
• ML-Ready Human Protein Data (2015-2023)
• High-Resolution Human Protein Structures with Ligands

Attributes