Functionally important positions can comprise the majority of a protein's architecture

Area of Science:

• Bioinformatics and Computational Biology
• Molecular Biology
• Protein Sequence Analysis

Background:

• Bioinformatics programs aim to identify functionally critical positions in protein sequences.
• Previous studies using the LacI/GalR family noted limitations in predicting all experimentally validated important positions, particularly in linker regions.

Purpose of the Study:

• To improve the identification of functionally important protein positions by refining sequence datasets and analysis strategies.
• To investigate whether knowledge-based sequence sorting enhances prediction accuracy for protein linker regions.

Main Methods:

• Corrected and expanded the commonly used LacI/GalR sequence set for analysis.
• Performed various bioinformatics analyses on the full and subsets of the LacI/GalR sequence data.
• Compared new predictions with existing experimental data for E. coli LacI and homologous linkers.

Main Results:

• Knowledge-based sequence sorting improved the identification of some linker positions.
• Despite improvements, two functionally important linker positions were consistently missed by all analyses.
• Estimated that over 50% of positions in LacI/GalR homologs are functionally important.

Conclusions:

• Current bioinformatics analyses have limitations in detecting all functionally critical protein positions, especially in linker regions.
• A significant proportion of protein positions appear to be functionally important, with neutral positions potentially being less common.
• Further experimental validation and understanding of sequence-function relationships are needed to fully decipher protein evolution and function.