Growth of novel protein structural data

Area of Science:

• Structural Biology
• Bioinformatics
• Computational Biology

Background:

• The exponential growth of structural data in the Protein Data Bank (PDB) has been a long-held assumption.
• Accurate measurement of novel protein structure growth is crucial for understanding trends in structural biology.

Purpose of the Study:

• To re-evaluate the growth rate of novel protein structural data in the PDB.
• To assess the impact of structural genomics (SG) initiatives on PDB growth and data redundancy.

Main Methods:

• Clustering protein entry sequences to identify novel structures.
• Developing a novel index to down-weight sequence-redundant entries.
• Analyzing the correlation between PDB growth and the expansion of Structural Classification of Proteins (SCOP) categories.

Main Results:

• The PDB's growth rate has slowed and is not exponential, a finding consistent across different measurement methods (PDB files, clusters, weighted chains).
• The expansion of SCOP categories is slowing, allowing for prediction of their final sizes.
• Structural genomics initiatives, particularly the NIH Protein Structure Initiative, have become major contributors since mid-2004, supplying half of the novel structures and yielding data that is 3.8 times less sequence-redundant.

Conclusions:

• The assumption of exponential PDB growth is incorrect; growth has decelerated.
• Structural genomics is vital for replenishing novel structural data and improving data quality in the PDB.
• Automated analysis provides an accurate and up-to-date measure of novel protein structural data growth.