Family-level specialization in protein domain insertion architectures

Area of Science:

• * Structural biology and bioinformatics
• * Protein domain architecture analysis

Background:

• * Multidomain proteins exhibit diverse architectures.
• * Domain insertion, where one domain interrupts another's sequence, is a key mechanism in protein evolution.

Purpose of the Study:

• * To comprehensively analyze domain insertion events across a large protein domain dataset.
• * To characterize the roles, size relationships, and structural context of inserted domains.
• * To provide a quantitative dataset for protein structure prediction and design.

Main Methods:

• * Analysis of 2.7 million classified protein domains.
• * Identification and classification of domain insertion events and protein families.
• * Quantitative analysis of domain sizes, roles, and structural superfamily relationships.

Main Results:

• * Domain insertions occur in 20% of multidomain proteins, involving 5701 families and 48,551 events.
• * 331 families show consistent roles as either hosts or inserted modules; 1116 families exhibit versatile roles.
• * Inserted domains are smaller than hosts, and insertions often bridge different structural superfamilies.

Conclusions:

• * Domain insertion is a significant driver of protein architecture diversity.
• * Consistent and versatile roles of inserted domains provide insights into protein evolution.
• * The comprehensive dataset and quantitative characterizations can advance protein structure prediction and design.