PISCES

PISCES filters and culls protein structures from the Protein Data Bank (PDB) to produce non-redundant, resolution-filtered sets of chains and entries for structural and bioinformatic analyses.

Key Features:

• Sequence Identity and Structural Quality Filtering: Filters sequences by mutual sequence identity thresholds and structural resolution quality to produce curated high-fidelity structure lists.
• By-Entry and By-Chain Culling: Implements culling by individual chains or entire PDB entries, with by-entry culling preventing any chain in an entry from exceeding the sequence identity threshold with chains in other entries.
• Integration with RCSB PDB Search: Accepts RCSB Protein Data Bank search results for refinement and culling using annotations and functional criteria.
• Annotated Sequences: Outputs sequences annotated with gene names and species information.
• Customizable Input Lists: Accepts custom lists of entries or chains to apply specified culling criteria.
• Regular Updates: Maintains data updated weekly to reflect current PDB content.

Scientific Applications:

• Homology Modeling: Provides template sets with low mutual sequence identity for selection of non-redundant modeling templates.
• Structural Comparisons: Generates non-redundant structure sets to support comparative structural analyses and similarity studies.
• Functional Studies: Enables selection based on functional annotations to support targeted studies of protein function and interaction networks.

Methodology:

PISCES uses PSI-BLAST and structure-based alignments to determine sequence identities, with structure alignments producing more complete alignments for precise identification of sequence similarities.