An analysis of structural instances of low complexity sequence segments

Area of Science:

• * Structural bioinformatics
• * Protein sequence analysis
• * Computational biology

Background:

• * Amino acid sequence databases reveal numerous low complexity, compositionally biased segments.
• * Few of these segments appear as short instances in proteins with known structures.
• * Understanding the structural characteristics of these segments is crucial.

Purpose of the Study:

• * To analyze structural instances of low complexity sequence segments in the Protein Data Bank.
• * To investigate preferences for sequence composition, secondary structure, and local atomic environment.
• * To determine if these segments are disordered or possess unique structural properties.

Main Methods:

• * Analysis of low complexity sequence segments within the Brookhaven Protein Data Bank.
• * Examination of sequence composition, secondary structural conformation, and local atomic environment.
• * Comparison of structural properties with predictions from secondary structure prediction methods.

Main Results:

• * Low complexity segments exhibit a near-linear relationship between complexity and length, indicating a lack of very long instances.
• * Identified segments are not disordered, with temperature factors comparable to the rest of the protein.
• * These segments are predominantly exposed and show a preference for helical or coiled conformations over random chance.

Conclusions:

• * Low complexity segments in proteins are structurally stable and often exposed.
• * Helical low complexity segments are well-predicted by secondary structure prediction tools.
• * Strand-forming low complexity segments present a challenge for current prediction methods.